Tissue engineering is a branch of regenerative medicine that harnesses biomaterial and stem cell research to utilise the body's natural healing responses to regenerate tissue and organs. There remain many unanswered questions in tissue engineering, with optimal biomaterial designs still to be developed and a lack of adequate stem cell knowledge limiting successful application. Advances in artificial intelligence (AI), and deep learning specifically, offer the potential to improve both scientific understanding and clinical outcomes in regenerative medicine. With enhanced perception of how to integrate artificial intelligence into current research and clinical practice, AI offers an invaluable tool to improve patient outcome.
Institute of Physics and Engineering in Medicine
IPEM's aim is to promote the advancement of physics and engineering applied to medicine and biology for the public benefit. Its members are professionals working in healthcare, education, industry and research.
IPEM publishes scientific journals and books and organises conferences to disseminate knowledge and support members in their development. It sets and advises on standards for the practice, education and training of scientists and engineers working in healthcare to secure an effective and appropriate workforce.
Purpose-led Publishing is a coalition of three not-for-profit publishers in the field of physical sciences: AIP Publishing, the American Physical Society and IOP Publishing.
Together, as publishers that will always put purpose above profit, we have defined a set of industry standards that underpin high-quality, ethical scholarly communications.
We are proudly declaring that science is our only shareholder.
ISSN: 2057-1976
A broad, inclusive, rapid review journal devoted to publishing new research in all areas of biomedical engineering, biophysics and medical physics, with a special emphasis on interdisciplinary work between these fields.
Open all abstracts, in this tab
Benita S Mackay et al 2021 Biomed. Phys. Eng. Express 7 052002
Christiana Subaar et al 2024 Biomed. Phys. Eng. Express 10 035029
A lot of underdeveloped nations particularly in Africa struggle with cancer-related, deadly diseases. Particularly in women, the incidence of breast cancer is rising daily because of ignorance and delayed diagnosis. Only by correctly identifying and diagnosing cancer in its very early stages of development can be effectively treated. The classification of cancer can be accelerated and automated with the aid of computer-aided diagnosis and medical image analysis techniques. This research provides the use of transfer learning from a Residual Network 18 (ResNet18) and Residual Network 34 (ResNet34) architectures to detect breast cancer. The study examined how breast cancer can be identified in breast mammography pictures using transfer learning from ResNet18 and ResNet34, and developed a demo app for radiologists using the trained models with the best validation accuracy. 1, 200 datasets of breast x-ray mammography images from the National Radiological Society's (NRS) archives were employed in the study. The dataset was categorised as implant cancer negative, implant cancer positive, cancer negative and cancer positive in order to increase the consistency of x-ray mammography images classification and produce better features. For the multi-class classification of the images, the study gave an average accuracy for binary classification of benign or malignant cancer cases of 86.7% validation accuracy for ResNet34 and 92% validation accuracy for ResNet18. A prototype web application showcasing ResNet18 performance has been created. The acquired results show how transfer learning can improve the accuracy of breast cancer detection, providing invaluable assistance to medical professionals, particularly in an African scenario.
Ander Biguri et al 2016 Biomed. Phys. Eng. Express 2 055010
In this article the Tomographic Iterative GPU-based Reconstruction (TIGRE) Toolbox, a MATLAB/CUDA toolbox for fast and accurate 3D x-ray image reconstruction, is presented. One of the key features is the implementation of a wide variety of iterative algorithms as well as FDK, including a range of algorithms in the SART family, the Krylov subspace family and a range of methods using total variation regularization. Additionally, the toolbox has GPU-accelerated projection and back projection using the latest techniques and it has a modular design that facilitates the implementation of new algorithms. We present an overview of the structure and techniques used in the creation of the toolbox, together with two usage examples. The TIGRE Toolbox is released under an open source licence, encouraging people to contribute.
Renata Saha et al 2024 Biomed. Phys. Eng. Express 10 035028
To treat diseases associated with vagal nerve control of peripheral organs, it is necessary to selectively activate efferent and afferent fibers in the vagus. As a result of the nerve's complex anatomy, fiber-specific activation proves challenging. Spatially selective neuromodulation using micromagnetic stimulation(μMS) is showing incredible promise. This neuromodulation technique uses microcoils(μcoils) to generate magnetic fields by powering them with a time-varying current. Following the principles of Faraday's law of induction, a highly directional electric field is induced in the nerve from the magnetic field. In this study on rodent cervical vagus, a solenoidal μcoil was oriented at an angle to left and right branches of the nerve. The aim of this study was to measure changes in the mean arterial pressure (MAP) and heart rate (HR) following μMS of the vagus. The μcoils were powered by a single-cycle sinusoidal current varying in pulse widths(PW = 100, 500, and 1000 μsec) at a frequency of 20 Hz. Under the influence of isoflurane, μMS of the left vagus at 1000 μsec PW led to an average drop in MAP of 16.75 mmHg(n = 7). In contrast, μMS of the right vagus under isoflurane resulted in an average drop of 11.93 mmHg in the MAP(n = 7). Surprisingly, there were no changes in HR to either right or left vagal μMS suggesting the drop in MAP associated with vagus μMS was the result of stimulation of afferent, but not efferent fibers. In urethane anesthetized rats, no changes in either MAP or HR were observed upon μMS of the right or left vagus(n = 3). These findings suggest the choice of anesthesia plays a key role in determining the efficacy of μMS on the vagal nerve. Absence of HR modulation upon μMS could offer alternative treatment options using VNS with fewer heart-related side-effects.
Taisa Higino and Rodrigo França 2022 Biomed. Phys. Eng. Express 8 042001
The use of nanoparticles as biomaterials with applications in the biomedical field is growing every day. These nanomaterials can be used as contrast imaging agents, combination therapy agents, and targeted delivery systems in medicine and dentistry. Usually, nanoparticles are found as synthetic or natural organic materials, such as hydroxyapatite, polymers, and lipids. Besides that, they are could also be inorganic, for instance, metallic or metal-oxide-based particles. These inorganic nanoparticles could additionally present magnetic properties, such as superparamagnetic iron oxide nanoparticles. The use of nanoparticles as drug delivery agents has many advantages, for they help diminish toxicity effects in the body since the drug dose reduces significantly, increases drugs biocompatibility, and helps target drugs to specific organs. As targeted-delivery agents, one of the applications uses nanoparticles as drug delivery particles for bone-tissue to treat cancer, osteoporosis, bone diseases, and dental treatments such as periodontitis. Their application as drug delivery agents requires a good comprehension of the nanoparticle properties and composition, alongside their synthesis and drug attachment characteristics. Properties such as size, shape, core-shell designs, and magnetic characteristics can influence their behavior inside the human body and modify magnetic properties in the case of magnetic nanoparticles. Based on that, many different studies have modified the synthesis methods for these nanoparticles and developed composite systems for therapeutics delivery, adapting, and improving magnetic properties, shell-core designs, and particle size and nanosystems characteristics. This review presents the most recent studies that have been presented with different nanoparticle types and structures for bone and dental drug delivery.
Nadia Muhammad Hussain et al 2024 Biomed. Phys. Eng. Express 10 022002
Intrapartum fetal hypoxia is related to long-term morbidity and mortality of the fetus and the mother. Fetal surveillance is extremely important to minimize the adverse outcomes arising from fetal hypoxia during labour. Several methods have been used in current clinical practice to monitor fetal well-being. For instance, biophysical technologies including cardiotocography, ST-analysis adjunct to cardiotocography, and Doppler ultrasound are used for intrapartum fetal monitoring. However, these technologies result in a high false-positive rate and increased obstetric interventions during labour. Alternatively, biochemical-based technologies including fetal scalp blood sampling and fetal pulse oximetry are used to identify metabolic acidosis and oxygen deprivation resulting from fetal hypoxia. These technologies neither improve clinical outcomes nor reduce unnecessary interventions during labour. Also, there is a need to link the physiological changes during fetal hypoxia to fetal monitoring technologies. The objective of this article is to assess the clinical background of fetal hypoxia and to review existing monitoring technologies for the detection and monitoring of fetal hypoxia. A comprehensive review has been made to predict fetal hypoxia using computational and machine-learning algorithms. The detection of more specific biomarkers or new sensing technologies is also reviewed which may help in the enhancement of the reliability of continuous fetal monitoring and may result in the accurate detection of intrapartum fetal hypoxia.
Li Liu et al 2018 Biomed. Phys. Eng. Express 4 015004
Adhesives that involve adhesion to the skin have been of great technological importance in medical or pharmaceutical fields, including recently emerging wearable sensors and electronics. The objective of this work was to evaluate the performances of silicone-based adhesives with skin, using a peel adhesion test. Specifically, we explored the effect of adhesive cleansing, which is an inevitable daily event for patients' comfort in long-term applications. Firstly, three medical grade silicone gels, Silbione® RT 4717, Silbione® RT 4642, and Silpuran® 2130, were used to fabricate adhesive pads. Their peel strength values were subsequently measured and compared, among which Silbione® RT gel 4717 possessed the highest peel strength. Therefore, it was selected as the raw material to fabricate the pads with a thickness range of 640–740 μm. Secondly, the peel adhesion of Silbione® RT 4717 adhesive pad was further compared with a series of commercial products that employ various medical-grade adhesives. The peel strength results indicated that our custom-made adhesive pad had an adequately strong adhesion for clinical use. Thirdly, in order to observe and predict the long-term performance of the adhesives, an aging test was performed in an ambient environment, revealing that Silbione® RT 4717 adhesive remained highly sticky for 5 days. Lastly, adequate cleansing protocols were established by monitoring the changes in peel strength after washing and wiping events. The reusability analysis showed that Silbione® 4717 adhesive pad was reusable in a one-week period for the washing method and 3 days for the wiping method.
Sotiris Raptis et al 2024 Biomed. Phys. Eng. Express 10 035016
Radiomics-based prediction models have shown promise in predicting Radiation Pneumonitis (RP), a common adverse outcome of chest irradiation. Τhis study looks into more than just RP: it also investigates a bigger shift in the way radiomics-based models work. By integrating multi-modal radiomic data, which includes a wide range of variables collected from medical images including cutting-edge PET/CT imaging, we have developed predictive models that capture the intricate nature of illness progression. Radiomic features were extracted using PyRadiomics, encompassing intensity, texture, and shape measures. The high-dimensional dataset formed the basis for our predictive models, primarily Gradient Boosting Machines (GBM)—XGBoost, LightGBM, and CatBoost. Performance evaluation metrics, including Multi-Modal AUC-ROC, Sensitivity, Specificity, and F1-Score, underscore the superiority of the Deep Neural Network (DNN) model. The DNN achieved a remarkable Multi-Modal AUC-ROC of 0.90, indicating superior discriminatory power. Sensitivity and specificity values of 0.85 and 0.91, respectively, highlight its effectiveness in detecting positive occurrences while accurately identifying negatives. External validation datasets, comprising retrospective patient data and a heterogeneous patient population, validate the robustness and generalizability of our models. The focus of our study is the application of sophisticated model interpretability methods, namely SHAP (SHapley Additive exPlanations) and LIME (Local Interpretable Model-Agnostic Explanations), to improve the clarity and understanding of predictions. These methods allow clinicians to visualize the effects of features and provide localized explanations for every prediction, enhancing the comprehensibility of the model. This strengthens trust and collaboration between computational technologies and medical competence. The integration of data-driven analytics and medical domain expertise represents a significant shift in the profession, advancing us from analyzing pixel-level information to gaining valuable prognostic insights.
Yunfei Hu et al 2022 Biomed. Phys. Eng. Express 8 025023
In this study, the performance of a new iterative reconstruction algorithm, the pre-clinical AcurosXB iCBCT algorithm, has been characterized on Varian Halcyon linear accelerators with respect to the potential of radiotherapy dose calculations on CBCT images. The study utilized various phantom setups to verify the accuracy of the pre-clinical algorithm under different scatter conditions and compared dose calculations performed on CBCT images reconstructed with the pre-clinical algorithm to those performed on typical planning CT images. The results indicated that despite showing improvements compared to the existing iCBCT protocol, certain restrictions should be introduced when the pre-clinical AcurosXB iCBCT algorithm was used for dose calculations. Changes in the scatter condition exhibited a larger effect on CBCTs than on planning CTs. Therefore, users should be careful in offsetting the patient and positioning the patient's arms if the resultant images will be used for dose calculations. In addition, protocols with different kV settings should be approached with caution, where 100 kV protocols should only be used to scan the head and neck area, while the rest of the body should be scanned with the 125 kV and 140 kV protocols. When the patient is set up properly and the appropriate energy is selected for the anatomical area, the uncertainty of using the novel AcurosXB iCBCT algorithm for treatment planning dose calculation is within ±2.0%.
Steve Collins et al 2024 Biomed. Phys. Eng. Express 10 035031
Background. Modern radiation therapy technologies aim to enhance radiation dose precision to the tumor and utilize hypofractionated treatment regimens. Verifying the dose distributions associated with these advanced radiation therapy treatments remains an active research area due to the complexity of delivery systems and the lack of suitable three-dimensional dosimetry tools. Gel dosimeters are a potential tool for measuring these complex dose distributions. A prototype tabletop solid-tank fan-beam optical CT scanner for readout of gel dosimeters was recently developed. This scanner does not have a straight raypath from source to detector, thus images cannot be reconstructed using filtered backprojection (FBP) and iterative techniques are required. Purpose. To compare a subset of the top performing algorithms in terms of image quality and quantitatively determine the optimal algorithm while accounting for refraction within the optical CT system. The following algorithms were compared: Landweber, superiorized Landweber with the fast gradient projection perturbation routine (S-LAND-FGP), the fast iterative shrinkage/thresholding algorithm with total variation penalty term (FISTA-TV), a monotone version of FISTA-TV (MFISTA-TV), superiorized conjugate gradient with the nonascending perturbation routine (S-CG-NA), superiorized conjugate gradient with the fast gradient projection perturbation routine (S-CG-FGP), superiorized conjugate gradient with with two iterations of CG performed on the current iterate and the nonascending perturbation routine (S-CG-2-NA). Methods. A ray tracing simulator was developed to track the path of light rays as they traverse the different mediums of the optical CT scanner. Two clinical phantoms and several synthetic phantoms were produced and used to evaluate the reconstruction techniques under known conditions. Reconstructed images were analyzed in terms of spatial resolution, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), signal non-uniformity (SNU), mean relative difference (MRD) and reconstruction time. We developed an image quality based method to find the optimal stopping iteration window for each algorithm. Imaging data from the prototype optical CT scanner was reconstructed and analysed to determine the optimal algorithm for this application. Results. The optimal algorithms found through the quantitative scoring metric were FISTA-TV and S-CG-2-NA. MFISTA-TV was found to behave almost identically to FISTA-TV however MFISTA-TV was unable to resolve some of the synthetic phantoms. S-CG-NA showed extreme fluctuations in the SNR and CNR values. S-CG-FGP had large fluctuations in the SNR and CNR values and the algorithm has less noise reduction than FISTA-TV and worse spatial resolution than S-CG-2-NA. S-LAND-FGP had many of the same characteristics as FISTA-TV; high noise reduction and stability from over iterating. However, S-LAND-FGP has worse SNR, CNR and SNU values as well as longer reconstruction time. S-CG-2-NA has superior spatial resolution to all algorithms while still maintaining good noise reduction and is uniquely stable from over iterating. Conclusions. Both optimal algorithms (FISTA-TV and S-CG-2-NA) are stable from over iterating and have excellent edge detection with ESF MTF 50% values of 1.266 mm−1 and 0.992 mm−1. FISTA-TV had the greatest noise reduction with SNR, CNR and SNU values of 424, 434 and 0.91 × 10−4, respectively. However, low spatial resolution makes FISTA-TV only viable for large field dosimetry. S-CG-2-NA has better spatial resolution than FISTA-TV with PSF and LSF MTF 50% values of 1.581 mm−1 and 0.738 mm−1, but less noise reduction. S-CG-2-NA still maintains good SNR, CNR, and SNU values of 168, 158 and 1.13 × 10−4, respectively. Thus, S-CG-2-NA is a well rounded reconstruction algorithm that would be the preferable choice for small field dosimetry.
Open all abstracts, in this tab
Khadidja Benchaira and Salim Bitam 2024 Biomed. Phys. Eng. Express 10 045010
Rapid and accurate electrocardiogram (ECG) signal classification is crucial in high-stakes healthcare settings. However, existing computational models often struggle to balance high performance with computational efficiency. This study introduces an innovative computational framework that combines transfer learning with traditional machine learning to optimize ECG classification. We use a pre-trained Stacked Convolutional Neural Network (SCNN) to generate high-dimensional feature embeddings, which are then evaluated by an array of machine learning classifiers. Our models demonstrate exceptional performance, particularly when utilizing embeddings from SCNNs trained on diverse datasets. This underscores the importance of data diversity in improving classifier discrimination. Notably, Multilayer Perceptrons (MLPs) stand out for their ability to balance computational efficiency with strong performance, achieving test F1-scores of 0.94 and 1.00 in multi-class and binary tasks on the CinC2017 dataset, and 0.85 and 0.99 on the CPSC2018 dataset. Our approach consistently outperforms existing methods, setting new benchmarks in ECG classification. The synergy between deep learning-based feature extraction and traditional machine learning through transfer learning offers a robust, efficient, and adaptable strategy for ECG classification, addressing a critical research gap and laying the groundwork for future advancements in this crucial healthcare field.
2024 Biomed. Phys. Eng. Express 10 040201
Luis Ángel Quiñones et al 2024 Biomed. Phys. Eng. Express 10 045009
Background. New applications of 3D printing have recently appeared in the fields of radiotherapy and radiology, but the knowledge of many radiological characteristics of the compounds involved is still limited. Therefore, studies are needed to improve our understanding about the transport and interaction of ionizing radiation in these materials. Purpose. The purpose of this study is to perform an analysis of the most important radiation interaction parameters in thermoplastic materials used in Fused Deposition Modeling 3D printing. Additionally, we propose improvements to bring their characteristics closer to those of water and use them as water substitutes in applications such as radiodiagnosis, external radiotherapy, and brachytherapy. Methods. We have calculated different magnitudes as mass linear attenuation, mass energy absorption coefficients, as well as stopping power and electronic density of several thermoplastic materials along with various compounds that have been used as water substitutes and in a new proposed blend. To perform these computations, we have used the XCOM and ESTAR databases from NIST and the EGSnrc code for Montecarlo simulations. Results. From the representation of the calculated interaction parameters, we have been able to establish relationships between their properties and the proportion of certain chemical elements. In addition, studying these same characteristics in different commercial solutions used as substitutes for water phantoms allows us to extrapolate improvements for these polymers. Conclusion. The radiological characteristics of the analyzed thermoplastic materials can be improved by adding some chemical elements with atomic numbers higher than oxygen and by using polyethylene in new blends.
Runhuang Yang et al 2024 Biomed. Phys. Eng. Express 10 045008
Objectives. Current lung cancer screening protocols primarily evaluate pulmonary nodules, yet often neglect the malignancy risk associated with small nodules (≤10 mm). This study endeavors to optimize the management of pulmonary nodules in this population by devising and externally validating a Multimodal Integrated Feature Neural Network (MIFNN). We hypothesize that the fusion of deep learning algorithms with morphological nodule features will significantly enhance diagnostic accuracy. Materials and Methods. Data were retrospectively collected from the Lung Nodule Analysis 2016 (LUNA16) dataset and four local centers in Beijing, China. The study includes patients with small pulmonary nodules (≤10 mm). We developed a neural network, termed MIFNN, that synergistically combines computed tomography (CT) images and morphological characteristics of pulmonary nodules. The network is designed to acquire clinically relevant deep learning features, thereby elevating the diagnostic accuracy of existing models. Importantly, the network's simple architecture and use of standard screening variables enable seamless integration into standard lung cancer screening protocols. Results. In summary, the study analyzed a total of 382 small pulmonary nodules (85 malignant) from the LUNA16 dataset and 101 small pulmonary nodules (33 malignant) obtained from four specialized centers in Beijing, China, for model training and external validation. Both internal and external validation metrics indicate that the MIFNN significantly surpasses extant state-of-the-art models, achieving an internal area under the curve (AUC) of 0.890 (95% CI: 0.848–0.932) and an external AUC of 0.843 (95% CI: 0.784–0.891). Conclusion. The MIFNN model significantly enhances the diagnostic accuracy of small pulmonary nodules, outperforming existing benchmarks by Zhang et al with a 6.34% improvement for nodules less than 10 mm. Leveraging advanced integration techniques for imaging and clinical data, MIFNN increases the efficiency of lung cancer screenings and optimizes nodule management, potentially reducing false positives and unnecessary biopsies. Clinical relevance statement. The MIFNN enhances lung cancer screening efficiency and patient management for small pulmonary nodules, while seamlessly integrating into existing workflows due to its reliance on standard screening variables.
S Senthilnathan et al 2024 Biomed. Phys. Eng. Express 10 045007
Cardiac electrical changes associated with ischemic heart disease (IHD) are subtle and could be detected even in rest condition in magnetocardiography (MCG) which measures weak cardiac magnetic fields. Cardiac features that are derived from MCG recorded from multiple locations on the chest of subjects and some conventional time domain indices are widely used in Machine learning (ML) classifiers to objectively distinguish IHD and control subjects. Most of the earlier studies have employed features that are derived from signal-averaged cardiac beats and have ignored inter-beat information. The present study demonstrates the utility of beat-by-beat features to be useful in classifying IHD subjects (n = 23) and healthy controls (n = 75) in 37-channel MCG data taken under rest condition of subjects. The study reveals the importance of three features (out of eight measured features) namely, the field map angle (FMA) computed from magnetic field map, beat-by-beat variations of alpha angle in the ST-T region and T wave magnitude variations in yielding a better classification accuracy (92.7 %) against that achieved by conventional features (81 %). Further, beat-by-beat features are also found to augment the accuracy in classifying myocardial infarction (MI) Versus control subjects in two public ECG databases (92 % from 88 % and 94 % from 77 %). These demonstrations summarily suggest the importance of beat-by-beat features in clinical diagnosis of ischemia.
Open all abstracts, in this tab
Mohammed Ali et al 2024 Biomed. Phys. Eng. Express 10 032003
Guided tissue/bone regeneration (GTR/GBR) is a widely used technique in dentistry to facilitate the regeneration of damaged bone and tissue, which involves guiding materials that eventually degrade, allowing newly created tissue to take its place. This comprehensive review the evolution of biomaterials for guided bone regeneration that showcases a progressive shift from non-resorbable to highly biocompatible and bioactive materials, allowing for more effective and predictable bone regeneration. The evolution of biomaterials for guided bone regeneration GTR/GBR has marked a significant progression in regenerative dentistry and maxillofacial surgery. Biomaterials used in GBR have evolved over time to enhance biocompatibility, bioactivity, and efficacy in promoting bone growth and integration. This review also probes into several promising fabrication techniques like electrospinning and latest 3D printing fabrication techniques, which have shown potential in enhancing tissue and bone regeneration processes. Further, the challenges and future direction of GTR/GBR are explored and discussed.
Muhammad Suhaib Shahid et al 2024 Biomed. Phys. Eng. Express 10 032001
The paper aims to explore the current state of understanding surrounding in silico oral modelling. This involves exploring methodologies, technologies and approaches pertaining to the modelling of the whole oral cavity; both internally and externally visible structures that may be relevant or appropriate to oral actions. Such a model could be referred to as a 'complete model' which includes consideration of a full set of facial features (i.e. not only mouth) as well as synergistic stimuli such as audio and facial thermal data. 3D modelling technologies capable of accurately and efficiently capturing a complete representation of the mouth for an individual have broad applications in the study of oral actions, due to their cost-effectiveness and time efficiency. This review delves into the field of clinical phonetics to classify oral actions pertaining to both speech and non-speech movements, identifying how the various vocal organs play a role in the articulatory and masticatory process. Vitaly, it provides a summation of 12 articulatory recording methods, forming a tool to be used by researchers in identifying which method of recording is appropriate for their work. After addressing the cost and resource-intensive limitations of existing methods, a new system of modelling is proposed that leverages external to internal correlation modelling techniques to create a more efficient models of the oral cavity. The vision is that the outcomes will be applicable to a broad spectrum of oral functions related to physiology, health and wellbeing, including speech, oral processing of foods as well as dental health. The applications may span from speech correction, designing foods for the aging population, whilst in the dental field we would be able to gain information about patient's oral actions that would become part of creating a personalised dental treatment plan.
Abdallah El Ouaridi et al 2024 Biomed. Phys. Eng. Express 10 032002
Positron emission tomography (PET) is a powerful medical imaging modality used in nuclear medicine to diagnose and monitor various clinical diseases in patients. It is more sensitive and produces a highly quantitative mapping of the three-dimensional biodistribution of positron-emitting radiotracers inside the human body. The underlying technology is constantly evolving, and recent advances in detection instrumentation and PET scanner design have significantly improved the medical diagnosis capabilities of this imaging modality, making it more efficient and opening the way to broader, innovative, and promising clinical applications. Some significant achievements related to detection instrumentation include introducing new scintillators and photodetectors as well as developing innovative detector designs and coupling configurations. Other advances in scanner design include moving towards a cylindrical geometry, 3D acquisition mode, and the trend towards a wider axial field of view and a shorter diameter. Further research on PET camera instrumentation and design will be required to advance this technology by improving its performance and extending its clinical applications while optimising radiation dose, image acquisition time, and manufacturing cost. This article comprehensively reviews the various parameters of detection instrumentation and PET system design. Firstly, an overview of the historical innovation of the PET system has been presented, focusing on instrumental technology. Secondly, we have characterised the main performance parameters of current clinical PET and detailed recent instrumental innovations and trends that affect these performances and clinical practice. Finally, prospects for this medical imaging modality are presented and discussed. This overview of the PET system's instrumental parameters enables us to draw solid conclusions on achieving the best possible performance for the different needs of different clinical applications.
Nadia Muhammad Hussain et al 2024 Biomed. Phys. Eng. Express 10 022002
Intrapartum fetal hypoxia is related to long-term morbidity and mortality of the fetus and the mother. Fetal surveillance is extremely important to minimize the adverse outcomes arising from fetal hypoxia during labour. Several methods have been used in current clinical practice to monitor fetal well-being. For instance, biophysical technologies including cardiotocography, ST-analysis adjunct to cardiotocography, and Doppler ultrasound are used for intrapartum fetal monitoring. However, these technologies result in a high false-positive rate and increased obstetric interventions during labour. Alternatively, biochemical-based technologies including fetal scalp blood sampling and fetal pulse oximetry are used to identify metabolic acidosis and oxygen deprivation resulting from fetal hypoxia. These technologies neither improve clinical outcomes nor reduce unnecessary interventions during labour. Also, there is a need to link the physiological changes during fetal hypoxia to fetal monitoring technologies. The objective of this article is to assess the clinical background of fetal hypoxia and to review existing monitoring technologies for the detection and monitoring of fetal hypoxia. A comprehensive review has been made to predict fetal hypoxia using computational and machine-learning algorithms. The detection of more specific biomarkers or new sensing technologies is also reviewed which may help in the enhancement of the reliability of continuous fetal monitoring and may result in the accurate detection of intrapartum fetal hypoxia.
Wei-Jen Chan and Huatian Li 2024 Biomed. Phys. Eng. Express 10 022001
In recent years, nanoparticles (NPs) have been extensively developed as drug carriers to overcome the limitations of cancer therapeutics. However, there are several biological barriers to nanomedicines, which include the lack of stability in circulation, limited target specificity, low penetration into tumors and insufficient cellular uptake, restricting the active targeting toward tumors of nanomedicines. To address these challenges, a variety of promising strategies were developed recently, as they can be designed to improve NP accumulation and penetration in tumor tissues, circulation stability, tumor targeting, and intracellular uptake. In this Review, we summarized nanomaterials developed in recent three years that could be utilized to improve drug delivery for cancer treatments.
Open all abstracts, in this tab
Manfred Köller et al 2018 Biomed. Phys. Eng. Express 4 055002
The role of bacterial cell division on the damage of adherent bacteria to titanium (Ti) nano-pillar cicada wing like surface was analyzed. Therefore nano-pillar Ti thin films were fabricated by glancing angle sputter deposition (GLAD) on silicon substrates. Gram-negative E. coli bacteria were allowed to adhere and to proliferate on these nanostructured samples for 3 h at 37 °C either under optimal cell growth conditions (brain heart infusion medium, BHI) or limited growth conditions (RPMI1640 medium). The bacteria adhered to the samples in both media. Compared to BHI medium the growth of E. coli in RPMI1640 medium was significantly inhibited. Concomitantly, the ratio of dead/living adherent bacteria on the nano-pillar surface was significantly decreased after the incubation period in RPMI1640. In addition, when the bacterial proliferation was biochemically halted using DL-serine-hydroxamate a comparable decrease in the ratio of dead/living adherent bacteria was also obtained in BHI medium. These results indicate that cell growth of adherent E. coli which is accompanied by cell elongations of the rod structure is involved in the damage induced by the titanium nano-pillar surface.
James Archer et al 2018 Biomed. Phys. Eng. Express 4 044003
Cherenkov radiation is the primary source of unwanted light in a scintillator dosimetry system. In this work we compare two techniques for temporally separating Cherenkov radiation from a slow scintillator signal. These techniques are applicable to a pulsed radiation beam. We found that by analysing the rising edge of the light pulse to identify the fast Cherenkov light only removed 74% of the Cherenkov light. By integrating the tail of the signal where only scintillation light is present a more accurate result is achieved. The average of the results of the two methods provides up to a 90% improvement in the accuracy of the relative dose when compared to ionisation chamber, in certain measurements. This work demonstrates an alternative methodology for the removal of Cherenkov light using signal analysis, while preserving all the scintillation light signal and minimising the bulk of the experimental equipment.
Natasha Maurmann et al 2017 Biomed. Phys. Eng. Express 3 045005
Materials, such as biopolymers, can be applied to produce scaffolds as mechanical support for cell growth in regenerative medicine. Two examples are polycaprolactone (PCL) and poly (lactic-co-glycolic acid) (PLGA), both used in this study to evaluate the behavior of umbilical cord-derived mesenchymal stem cells. The scaffolds were produced by the 3D printing technique using PCL as a polymer covered with PLGA fibers obtained by electrospinning. The cells were seeded in three concentrations: 8.5 × 103; 25.5 × 103 and 51.0 × 103 on the two surfaces of the scaffolds. With scanning electron microscopy (SEM), it was observed that the electrospun fibers were integrated into the 3D printed matrices. Confocal laser scanning microscopy and SEM confirmed the presence of attached cells and the lactate dehydrogenase release test showed the scaffolds were not cytotoxic. The cells were able to differentiate into osteogenic and chondrogenic lineages on the scaffolds. Mechanical test showed that the cells seeded on the 3D printed PCL matrices coated with PLGA electrospun nanofibers (3D + ES + SC) did not show significant difference in tensile modulus than the pure PCL matrix (3D) or PCL matrices coated with PLGA electrospun nanofibers (3D + ES). The combination of the two polymers facilitated the production of a support with greater mechanical stability due to the presence of the 3D printed PCL matrices fabricated by melted filaments and greater cell adhesion due to the PLGA fibers. The scaffolds are suitable for use in cell therapy and also for tissue regeneration purposes.
Open all abstracts, in this tab
Parker et al
A study of burn thresholds from superficially penetrating radio-frequency (RF) energy at 8.2 and 95 GHz for swine skin was conducted. The study determined the thresholds for superficial, partial-thickness, and full-thickness burn severities after 5 seconds of exposure at power densities of 4–30 W/cm2 and 2–15~W/cm2 at 8.2 and 95 GHz, respectively. There were significant differences in the burn thresholds at the different severities between the two frequencies due to the large difference in energy penetration depths. Biopsies were collected from each burn site at 1, 24, 72, and 168 hr post-exposure. Each sample was assessed by a burn pathologist against 20 histological factors to characterize the damage resulting from these RF overexposures. A one-dimensional, layered digital phantom that utilized realistic values for dielectric and thermal properties was used to explain some observed thresholds. The results of the heating and cooling response of the animal model, and histology scores of each exposure are provided to enhance future efforts at simulation of RF overexposures and to establish damage thresholds.
Kang et al
Evaluation of skin recovery is an important step in the treatment of burns. However, conventional methods only observe the surface of the skin and cannot quantify the injury volume. Optical coherence tomography (OCT) is a non-invasive, non-contact, real-time technique. Swept-frequency OCT uses near infrared light and analyzes the intensity of light echo at different depths to generate images from optical interference signals. To quantify the dynamic recovery of skin burns over time, laser induced skin burns in mice were evaluated using deep learning of swept-frequency OCT images.A laser-induced mouse skin thermal injury model was established in thirty Kunming mice, and OCT images of normal and burned areas of mouse skin were acquired at day 0, day 1, day 3, day 7, and day 14 after laser irradiation. This resulted in 7000 normal and 1400 burn B-scan images which were divided into training, validation, and test sets at 8:1.5:0.5 ratio for the normal data and 8:1:1 for the burn data. Normal images were manually annotated, and the deep learning U-Net model (verified with PSPNe and HRNet models) was used to segment the skin into three layers: the dermal epidermal layer, subcutaneous fat layer, and muscle layer. For the burn images, the models were trained to segment just the damaged area. Three-dimensional reconstruction technology was then used to reconstruct the damaged tissue and calculate the damaged tissue volume. The average IoU value and f-score of the normal tissue layer U-Net segmentation model were 0.876 and 0.934 respectively. The IoU value of the burn area segmentation model reached 0.907 and f-score value reached 0.951. Compared with manual labeling, the U-Net model was faster with higher accuracy for skin stratification. OCT and U-Net segmentation can provide rapid and accurate analysis of tissue changes and clinical guidance in the treatment of burns.
Ara et al
Abstract
The Z-scan technique is a nonlinear optical method that has found applications in characterizing various materials, particularly those exhibiting nonlinear optical response (NLOR). This study applies the continuous wave (CW) Z-scan technique to examine the NLOR in terms of the nonlinear optical phase shifts (ΔΦ_0) exhibited by the ccfDNA extracted from blood plasma samples collected from a group constituting 30 cancer-diagnosed patients and another group constituting 30 non-diagnosed individuals. The cancer group exhibited significantly higher ΔΦ_0vs. incident power slopes compared to the non-cancer group (0.34 vs 0.12) providing a clear distinction between the two groups. The receiver operating characteristic (ROC) curve analysis of the results indicates a clear separation between cancer and non-cancer groups, along with a 94% accuracy rate of the data. The Z-scan results are corroborated by spectrophotometric analysis, revealing a consistent trend in the concentration values of ccfDNA samples extracted from both cancerous and non-cancerous samples, measuring 3.24 and 1.41 respectively. Additionally, more sensitive fluorometric analyses of the respective samples demonstrate significantly higher concentrations of ccfDNA in the cancer group, further affirming the correlation with the Z-scan results. The study suggests that the Z-scan technique holds promise as an effective method for cancer detection, potentially contributing to improved oncology diagnosis and prognosis in the future.
Chaitanya et al
Keratoprosthesis (KPro) is a surgical procedure largely confined to end-stage corneal blindness correction, where artificial cornea substitutes the native tissue. Though the problem of bio integration was addressed partially by strategic utilization of synthetic polymers and native tissue, major challenges like optical performance and design-associated post-operative complications of KPro were overlooked. Herein, a novel intralamellar KPro design is conceptualized to address these challenges using a light-transparent poly(2-hydroxy ethylmethacrylate) (pHEMA) hydrogel with good shape memory. pHEMA-based optics' theoretically modelled refractive surfaces for both phakic and aphakic conditions were investigated against the standard Navarro model and optimized to new aspheric geometries having high optical functionality utilizing the Zemax OpticStudio software. The optical clear aperture size standardized achieved a 15 % improvement in the illumination field. The introduction of asphericity on the two refractive surfaces of the optic on both models resulted in substantial improvements in the spot spread confinement on the retina, spatial resolution, and Seidel aberration. The design simulation study shows that the developed materials' optical characteristics coupled with newly optimized refractive surface geometries can indeed deliver very high visual performance. Furthermore, the procedure can be adapted to analyze and optimize the optical performance of a KPro, intraocular lens, or contact lens.
KANJIRATH EDDAM et al
The surface dose response of luminescent detectors (LD), in particular thermoluminescent dosimeter (TLD) and optically stimulated luminescent dosimeter (OSLD), during modified re constructive mastectomy (MRM) irradiation has been compared. The study utilized 100 MRM patients with LDs (n=500) for the irradiation of TLD (n=250) and OSLD (n=250) during irradiation from an Elekta Versa HD Linear accelerator.A more pronounced dose-response relationship was evident for both TLD and OSLD. These findings indicate a close adherence to the vendor-specified tolerance limits, affirming the suitability of these dosimeters, as substantiated by individual investigations on each type. While a minor difference in dose measurement was noted between TLD and OSLD, the study's outcomes support the applicability of both TLD and OSLD as effective dosimeters during MRM breast irradiation for surface dose evaluation.
Trending on Altmetric
Open all abstracts, in this tab
null 2024 Biomed. Phys. Eng. Express 10 040201
James E Parker et al 2024 Biomed. Phys. Eng. Express
A study of burn thresholds from superficially penetrating radio-frequency (RF) energy at 8.2 and 95 GHz for swine skin was conducted. The study determined the thresholds for superficial, partial-thickness, and full-thickness burn severities after 5 seconds of exposure at power densities of 4–30 W/cm2 and 2–15~W/cm2 at 8.2 and 95 GHz, respectively. There were significant differences in the burn thresholds at the different severities between the two frequencies due to the large difference in energy penetration depths. Biopsies were collected from each burn site at 1, 24, 72, and 168 hr post-exposure. Each sample was assessed by a burn pathologist against 20 histological factors to characterize the damage resulting from these RF overexposures. A one-dimensional, layered digital phantom that utilized realistic values for dielectric and thermal properties was used to explain some observed thresholds. The results of the heating and cooling response of the animal model, and histology scores of each exposure are provided to enhance future efforts at simulation of RF overexposures and to establish damage thresholds.
Camilla Scapicchio et al 2024 Biomed. Phys. Eng. Express 10 045006
Objective. Radiomics is a promising valuable analysis tool consisting in extracting quantitative information from medical images. However, the extracted radiomics features are too sensitive to variations in used image acquisition and reconstruction parameters. This limited robustness hinders the generalizable validity of radiomics-assisted models. Our aim is to investigate a possible harmonization strategy based on matching image quality to improve feature robustness. Approach. We acquired CT scans of a phantom with two scanners across different dose levels and percentages of Iterative Reconstruction algorithms. The detectability index was used as a comprehensive task-based image quality metric. A statistical analysis based on the Intraclass Correlation Coefficient was performed to determine if matching image quality/appearance could enhance the robustness of radiomics features extracted from the phantom images. Additionally, an Artificial Neural Network was trained on these features to automatically classify the scanner used for image acquisition. Main results. We found that the ICC of the features across protocols providing a similar detectability index improves with respect to the ICC of the features across protocols providing a different detectability index. This improvement was particularly noticeable in features relevant for distinguishing between scanners. Significance. This preliminary study demonstrates that a harmonization based on image quality/appearance matching could improve radiomics features robustness and heterogeneous protocols can be used to obtain a similar image appearance in terms of the detectability index. Thus protocols with a lower dose level could be selected to reduce the amount of radiation dose delivered to the patient and simultaneously obtain a more robust quantitative analysis.
Steven Squires et al 2024 Biomed. Phys. Eng. Express
Purpose: To improve breast cancer risk prediction for young women, we have developed deep learning methods to estimate mammographic density from low dose mammograms taken at approximately 1/10th of the usual dose. We investigate the quality and reliability of the density scores produced on low dose mammograms focussing on how image resolution and levels of training affect the low dose predictions.
Methods: Deep learning models are developed and tested, with two feature extraction methods and an end-to-end trained method, on five different resolutions of 15,290 standard dose and simulated low dose mammograms with known labels. The models are further tested on a dataset with 296 matching standard and real low dose images allowing performance on the low dose images to be ascertained.
Results: Prediction quality on standard and simulated low dose images compared to labels is similar for all equivalent model training and image resolution versions. Increasing resolution results in improved performance of both feature extraction methods for standard and simulated low dose images, while the trained models show high performance across the resolutions. For the trained models the Spearman rank correlation coefficient between predictions of standard and low dose images at low resolution is 0.951 (0.937 to 0.960) and at the highest resolution 0.956 (0.942 to 0.965). If pairs of model predictions are averaged, similarity increases.
Conclusions: Deep learning mammographic density predictions on low dose mammograms are highly correlated with standard dose equivalents for feature extraction and end-to-end approaches across multiple image resolutions. Deep learning models can reliably make high quality mammographic density predictions on low dose mammograms.
Esmaeil Mehrara 2024 Biomed. Phys. Eng. Express
Thermoluminescent dosimeters (TLDs) serve as compact and user-friendly tools for various applications, including personal radiation dosimetry and radiation therapy. This study explores the potential of utilizing TLD-100 personal dosimetry, conventionally applied in PET/CT (positron emission tomography/computed tomography) settings, in the PET/MRI (magnetic resonance imaging) environment. The integration of MRI into conventional radiotherapy and PET systems necessitates ionizing radiation dosimetry in the presence of static magnetic fields. In this study, TLD-100 dosimeters were exposed on the surface of a water-filled cylindrical phantom containing PET-radioisotope and positioned on the patient table of a 3 T PET/MRI, where the magnetic field strength is around 0.2 T, aiming to replicate real-world scenarios experienced by personnel in PET/MRI environments. . Results indicate that the modified MR-safe TLD-100 personal dosimeters exhibit no significant impact from the static magnetic field of the 3 T PET/MRI, supporting their suitability for personal dosimetry in PET/MRI settings. This study addresses a notable gap in existing literature on the effect of MRI static magnetic field on TLDs.
Sruthi Sivabhaskar et al 2024 Biomed. Phys. Eng. Express
Objective: The aim of this work was to develop a Phase I control chart framework for the recently proposed multivariate risk-adjusted Hotelling's T^2 chart. Although this control chart alone can identify most patients receiving extreme organ-at-risk (OAR) dose, it is restricted by underlying distributional assumptions, making it sensitive to extreme observations in the sample, as is typically found in radiotherapy plan quality data such as dose-volume histogram (DVH) points. This can lead to slightly poor-quality plans that should have been identified as out-of-control (OC) to be signaled in-control (IC). 
Approach: We develop a robust iterative control chart framework to identify all OC patients with abnormally high OAR dose and improve them via re-optimization to achieve an IC sample prior to establishing the Phase I control chart, which can be used to monitor future treatment plans. 
Main Results: Eighty head-and-neck patients were used in this study. After the first iteration, P14, P67, and P68 were detected as OC for high brainstem dose, warranting re-optimization aimed to reduce brainstem dose without worsening other planning criteria. The DVH and control chart were updated after re-optimization. On the second iteration, P14, P67, and P68 were IC, but P40 was identified as OC. After re-optimizing P40's plan and updating the DVH and control chart, P40 was IC, but P14* (P14's re-optimized plan) and P62 were flagged as OC. P14* could not be re-optimized without worsening target coverage, so only P62 was re-optimized. Ultimately, a fully IC sample was achieved. Multiple iterations were needed to identify and improve all OC patients, and to establish a more robust control limit to monitor future treatment plans.
Significance: The iterative procedure resulted in a fully IC sample of patients. With this sample, a more robust Phase I control chart that can monitor OAR doses of new plans was established.
Owen Paetkau et al 2024 Biomed. Phys. Eng. Express
Background and purpose: To investigate models developed using radiomic and dosiomic (multi-omics) features from planning and treatment imaging for late patient-reported dysphagia in head and neck radiotherapy.

Materials and methods: Training (n=64) and testing (n=23) cohorts of head and neck cancer patients treated with curative intent chemo-radiotherapy with a follow-up time greater than 12 months were retrospectively examined. Patients completed the MD Anderson Dysphagia Inventory and a composite score ≤60 was interpreted as patient-reported dysphagia. A chart review collected baseline dysphagia and clinical factors. Multi-omic features were extracted from planning and last synthetic CT images using the pharyngeal constrictor muscle contours as a region of interest. Late patient-reported dysphagia models were developed using a random forest backbone, with feature selection and up-sampling methods to account for the imbalanced data. Models were developed and validated for multi-omic feature combinations for both timepoints. 

Results: A clinical and radiomic feature model developed using the planning CT achieved good performance (validation: sensitivity=80±27% / balanced accuracy=71±23%, testing: sensitivity =80±10% / balanced accuracy=73±11%). The synthetic CT models did not show improvement over the plan CT multi-omics models, with poor reliability of the radiomic features on these images. Dosiomic features extracted from the synthetic CT showed promise in predicting late patient-reported dysphagia.

Conclusion: Multi-omics models can predict late patient-reported dysphagia in head and neck radiotherapy patients. Synthetic CT dosiomic features show promise in developing successful models to account for changes in delivered dose distribution. Multi-center or prospective studies are required prior to clinical implementation of these models.
Ma'Moun Abu-Ayyad et al 2024 Biomed. Phys. Eng. Express
Magnetic nanoparticle hyperthermia (MNPH) has emerged as a promising cancer treatment that complements the conventional ionizing radiation and chemotherapy. MNPH involves the injection of iron oxide nanoparticles into the tumor and exposure to an alternating magnetic field (AMF). Iron oxide nanoparticles generate heat due to hysteresis loss when exposed to radiofrequency AMF. Exposing human tissue to AMF causes non-specific heating in tissues through induced eddy currents, which must be minimized. A pulse-width-modulated AMF has been shown to minimize eddy current heating in superficial tissues. This project developed a control strategy based on a simplified mathematical model in MATLAB SIMULINK® to minimize eddy current heating while maintaining a therapeutic temperature in the tumor. A minimum tumor temperature of 43 [°C] tumor temperature is required for at least 30 [min] while maintaining the surrounding healthy tissues below 39 [°C]. A model predictive control (MPC) algorithm was used to reach the target temperature within approximately 100 [s]. As a constrained MPC approach, a maximum AMF amplitude of 36 [kA/m] and increment of 5 [kA/m/s] were applied. The MPC used the AMF amplitude as an input and the open-loop response of the eddy current heating in its dynamic matrix. A conventional proportional integral (PI) controller was implemented and compared to the MPC performance. The results showed that MPC had a faster response (30 [s]) with minimal overshoot (1.4 [%]) compared to PI (115 [s] and 5.7 [%]) responses. In addition, the MPC method performed better than the structured PI controller in its ability to handle constraints and changes in process parameters.
Wondesen T. Gebreamlak and Hassaan H. Alkhatib 2024 Biomed. Phys. Eng. Express
Purpose:
The purpose of this study is to compute the planar dose distribution of irregularly shaped electron beams at their maximum dose depth (zmax) using the modified lateral build-up-ratio (LBR) and curve-fitting methods.
Methods:
Circular and irregular cutouts were created using Cerrobend alloy for a 14×14 cm2 applicator. Percentage depth dose (PDD) at the standard source-surface-distance (SSD = 100 cm) and point dose at different SSD were measured for each cutout. Orthogonal profiles of the cutouts were measured at zmax. The data were collected for 6, 9, 12, and 15 MeV electron beam energies on a VERSA HDTM LINAC using the iba Blue Phantom2 3D water phantom system. The planar dose distributions of the cutouts were also measured at zmax in solid water using EDR2 films.
Results:
The measured PDD curves were normalized to a normalization depth (d0) of 1 mm. Each cutout's lateral-buildup-ratio (LBR), lateral spread parameter (σR(z)), and effective SSD (SSDeff) were calculated using the PDD of the open applicator as the reference field. The modified LBR method was then used to calculate the planar dose distribution of the irregular cutouts inside the field at least 5 mm from the edge. A simple curve-fitting model was developed based on the profile shapes of the circular cutouts around the field edge. This model was used to calculate the planar dose distribution of irregular cutouts in the region from 3 mm outside to 5 mm inside the field edge. Finally, the calculated dose distribution was compared with the film measurement. 
Conclusions:
The planar dose distribution of electron therapy for irregular cutouts at zmax was calculated using the improved LBR method and a simple curve-fitting model. The calculated profiles were within 3% of the measured. The gamma passing rate with 3%/3mm was more than 96%.
Fan Peng et al 2024 Biomed. Phys. Eng. Express 10 035038
Objective. Ultrasound-assisted orthopaedic navigation held promise due to its non-ionizing feature, portability, low cost, and real-time performance. To facilitate the applications, it was critical to have accurate and real-time bone surface segmentation. Nevertheless, the imaging artifacts and low signal-to-noise ratios in the tomographical B-mode ultrasound (B-US) images created substantial challenges in bone surface detection. In this study, we presented an end-to-end lightweight US bone segmentation network (UBS-Net) for bone surface detection. Approach. We presented an end-to-end lightweight UBS-Net for bone surface detection, using the U-Net structure as the base framework and a level set loss function for improved sensitivity to bone surface detectability. A dual attention (DA) mechanism was introduced at the end of the encoder, which considered both position and channel information to obtain the correlation between the position and channel dimensions of the feature map, where axial attention (AA) replaced the traditional self-attention (SA) mechanism in the position attention module for better computational efficiency. The position attention and channel attention (CA) were combined with a two-class fusion module for the DA map. The decoding module finally completed the bone surface detection. Main Results. As a result, a frame rate of 21 frames per second (fps) in detection were achieved. It outperformed the state-of-the-art method with higher segmentation accuracy (Dice similarity coefficient: 88.76% versus 87.22%) when applied the retrospective ultrasound (US) data from 11 volunteers. Significance. The proposed UBS-Net for bone surface detection in ultrasound achieved outstanding accuracy and real-time performance. The new method out-performed the state-of-the-art methods. It had potential in US-guided orthopaedic surgery applications.