Medical Engineering & Physics最新文献

{"title":"3-D contour-aware U-Net for efficient rectal tumor segmentation in magnetic resonance imaging","authors":"Yi Lu ,&nbsp;Jun Dang ,&nbsp;Junzhang Chen ,&nbsp;Yuanyuan Wang ,&nbsp;Tao Zhang ,&nbsp;Xiangzhi Bai","doi":"10.1016/j.medengphy.2025.104352","DOIUrl":"10.1016/j.medengphy.2025.104352","url":null,"abstract":"<div><div>Magnetic resonance imaging (MRI), as a non-invasive detection method, is crucial for the clinical diagnosis and treatment plan of rectal cancer. However, due to the low contrast of rectal tumor signal in MRI, segmentation is often inaccurate. In this paper, we propose a new three-dimensional rectal tumor segmentation method CAU-Net based on T2-weighted MRI images. The method adopts a convolutional neural network to extract multi-scale features from MRI images and uses a Contour-Aware decoder and attention fusion block (AFB) for contour enhancement. We also introduce adversarial constraint to improve augmentation performance. Furthermore, we construct a dataset of 108 MRI-T2 volumes for the segmentation of locally advanced rectal cancer. Finally, CAU-Net achieved a DSC of 0.7112 and an ASD of 2.4707, which outperforms other state-of-the-art methods. Various experiments on this dataset show that CAU-Net has high accuracy and efficiency in rectal tumor segmentation. In summary, proposed method has important clinical application value and can provide important support for medical image analysis and clinical treatment of rectal cancer. With further development and application, this method has the potential to improve the accuracy of rectal cancer diagnosis and treatment.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104352"},"PeriodicalIF":1.7,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143894920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient digitally reconstructed radiographs with middle-out rendering and convex hull-based early stopping","authors":"William Burton,&nbsp;Casey Myers ,&nbsp;Chadd Clary ,&nbsp;Paul Rullkoetter","doi":"10.1016/j.medengphy.2025.104351","DOIUrl":"10.1016/j.medengphy.2025.104351","url":null,"abstract":"<div><div>Kinematic tracking of native anatomy in dynamic radiography facilitates understanding of <em>in vivo</em> movement. Tracking is performed using model-image registration, which involves estimating 6 degree-of-freedom poses of anatomic structures by comparing captured radiographs to dynamically rendered images of digital models. This procedure can produce accurate pose estimates of native anatomy, but computational efficiency remains a concern. A key source of latency in model-image registration is the iterative rendering of digitally reconstructed radiographs, required for evaluation of candidate poses. This technical note introduces an efficient algorithm for rendering digitally reconstructed radiographs. The proposed method is shown to accelerate rendering speeds by up to 50% compared to a baseline method which is commonly used in registration frameworks. Efficient rendering of digitally reconstructed radiographs may enhance the overall viability of model-image registration for use in kinematic tracking, and may contribute to bringing this technology closer to adoption in clinical settings.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104351"},"PeriodicalIF":1.7,"publicationDate":"2025-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143894805","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fundus image based diabetic retinopathy detection using EfficientNetB3 with squeeze and excitation block","authors":"Ravi Bhushan Dixit,&nbsp;Chandan Kumar Jha","doi":"10.1016/j.medengphy.2025.104350","DOIUrl":"10.1016/j.medengphy.2025.104350","url":null,"abstract":"<div><div>Diabetic retinopathy (DR) is a retinal affliction in patients suffering from diabetes. If DR is unidentified at an earlier stage, it may lead to blindness. Manual screening of DR using fundus images is a complex and time-consuming task. In the past, many automated techniques have been developed for DR detection and classification. In the case of multiclass fundus images, producing reliable classification performance is a challenge for researchers. Hence, this paper presents a novel transfer learning-based approach to classify DR using fundus images. The proposed technique is based on EfficientNetB3 with squeeze and excitation block. EfficientNetB3 performs classification tasks very well using an effective architecture with fewer parameters while the squeeze and excitation block improves the model's ability by focusing on crucial features. For experimentation of the proposed technique, fundus images of the APTOS-2019 dataset are utilized. The proposed technique achieves overall 88.44% accuracy, 98.00% specificity, 84.00% precision, 83.00% sensitivity, 83.00% F1-score, and 0.88 kappa score for all five classes of fundus images of the APTOS-2019 dataset. In addition to this, the proposed technique is also experimented using fundus images of the IDRiD and Messidor-2 datasets. The performance of the proposed technique is better than many existing DR detection techniques.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104350"},"PeriodicalIF":1.7,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143886368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Phase changes of the flow rate in the vertebral artery caused by debranching thoracic endovascular aortic repair: Effects of flow path and local vessel stiffness on vertebral arterial pulsation","authors":"Naoki Takeishi ,&nbsp;Li Jialong ,&nbsp;Naoto Yokoyama ,&nbsp;Takasumi Goto ,&nbsp;Hisashi Tanaka ,&nbsp;Shigeru Miyagawa ,&nbsp;Shigeo Wada","doi":"10.1016/j.medengphy.2025.104348","DOIUrl":"10.1016/j.medengphy.2025.104348","url":null,"abstract":"<div><div>Despite numerous studies on cerebral arterial blood flow, there has not yet been a comprehensive description of hemodynamics in patients undergoing debranching thoracic endovascular aortic repair (dTEVAR), a promising surgical option for aortic arch aneurysms. A phase delay of the flow rate in the left vertebral artery (LVA) in patients after dTEVAR compared to those before was experimentally observed, while the phase in the right vertebral artery (RVA) remained almost the same before and after surgery. Since this surgical intervention included stent graft implantation and extra-anatomical bypass, it was expected that the intracranial hemodynamic changes due to dTEVAR were coupled with fluid flow and pulse waves in cerebral arteries. To clarify this issue, a one-dimensional model (1D) was used to numerically investigate the relative contribution (i.e., local vessel stiffness and flow path changes) of the VA flow rate to the phase difference. The numerical results demonstrated a phase delay of flow rate in the LVA but not the RVA in postoperative patients undergoing dTEVAR relative to preoperative patients. The results further showed that the primary factor affecting the phase delay of the flow rate in the LVA after surgery compared to that before was the bypass, i.e., alteration of flow path, rather than stent grafting, i.e., the change in local vessel stiffness. The numerical results provide insights into hemodynamics in postoperative patients undergoing dTEVAR, as well as knowledge about therapeutic decisions.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104348"},"PeriodicalIF":1.7,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143886367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Realistic approach for sub-bandage pressure appraisal","authors":"Ferdinand Tamoue ,&nbsp;Tomasz Blachowicz ,&nbsp;Gunnar Riepe","doi":"10.1016/j.medengphy.2025.104345","DOIUrl":"10.1016/j.medengphy.2025.104345","url":null,"abstract":"<div><div>Manufacturers estimate the specific compression levels of their products to satisfy the intended medical purposes. Medical compression bandages have different intended purposes. This means that the engineer must be able to estimate the clinical performance, referred to as the interface pressure, during product development and prior to clinical testing in humans. Unfortunately, the mathematical equations found in the literature do not result in interface pressure values that are similar to the ex vivo pressure. Therefore, the present work aims to demonstrate that the use of bandage elongation as a premise for predicting the interface pressure, as revealed new expression, can provide a reliable estimate.</div></div><div><h3>Methods</h3><div>Pressure of compression bandages can be discrepant when using different method or devices. This work proposes a modified equation of Pascal and Laplace, used to estimate the sub-bandage pressure, is given by: <em>p</em> = (2*π*F)/ (ε+1)*L0*w, where P, represents the pressure in kPa, F is the force applied on the bandage width (w), L0 is the initial length, and ε = (Δ_L/L0)*100 is the relative elongation expressed in [%].</div><div>Bandage specimens underwent a tensile test using a constant rate extension (CRE) device. The tensile test determined the elongation resulting from an arbitrary force and predicted the sub-bandage pressure. The elongation was then prescribed as the stretch required to achieve ex vivo pressure. The elongation and pressure were measured ex vivo with a ruler and the PicoPress® device, respectively, and compared to the predicted elongation and pressure.</div></div><div><h3>Result</h3><div>The predicted pressure (24 mm Hg), calculated for ε = 48 % in the pressure equation, is not significantly different from the ex vivo pressure (26 mm Hg), as measured by a healthcare professional using the PicoPress®. The p-value (0.102) is greater than α (0.05) and the Pearson coefficient, <em>R</em> = 0.529, indicates a moderate relationship between the two variables.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104345"},"PeriodicalIF":1.7,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143894919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dosimetric impact of spot size variations in proton pencil beam scanning — a Monte Carlo simulation study","authors":"Jiayi Guo ,&nbsp;Fuquan Zhang ,&nbsp;Jingyi Cheng ,&nbsp;Rong Zhou ,&nbsp;Yinxiangzi Sheng","doi":"10.1016/j.medengphy.2025.104346","DOIUrl":"10.1016/j.medengphy.2025.104346","url":null,"abstract":"<div><h3>Background and Purpose</h3><div>Spot size is a crucial parameter in the clinical operation of light-ion beam therapy delivery systems. This study aims to quantify the influence of spot size variations on dose distribution in proton pencil beam radiotherapy.</div></div><div><h3>Materials and Methods</h3><div>Three cubic targets with modulation widths (M) of 3 cm, 6 cm, and 9 cm were positioned at center depths (CD) of 5 cm, 15 cm, and 25 cm within a water phantom. Dosimetric evaluations including 3D γ passing rate (γ-PR), 2D point-to-point dose, lateral penumbra, and flatness, were performed using a Monte Carlo tool. Furthermore, an additional analysis was performed using 3D γ-PR to evaluate the spot size variations of three patients with varying tumor sites and volumes.</div></div><div><h3>Results</h3><div>With a tolerance of 2 % and 2 mm, the dose distribution changes slightly for a 10 % increase in spot size (3D γ-PR&gt;95 %). Decreasing the spot size by 10 %, the 3D γ-PR values were no &lt;90 %. Furthermore, as depth increases, the 3D γ-PR values tend to increase. The 2D point-to-point dose deviations remained below 3 % for spot size variations of ±10 %. Additionally, as depth increases, the 2D point-to-point dose deviations decrease. Increasing the spot size by 10 % led to a maximum change in lateral penumbra of 1.53 mm, while decreasing the spot size by 10 % resulted in lateral penumbra deviations consistently below 1 mm for all cube targets. Additionally, as depth increases, the absolute value of lateral penumbra deviations tends to decrease. With both a 10 % decrease and increase in spot size, the deviations of flatness were always lower than 3 %. And the deviations of flatness decrease as depth increases.</div></div><div><h3>Conclusions</h3><div>The decreasing and increasing spot size had different effects in terms of disturbing the target dose. The dosimetric impact of spot size variations was strongly correlated with target depth. Higher-energy beams showed less sensitivity to the perturbation. Thus, a higher spot size tolerance could be expected.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104346"},"PeriodicalIF":1.7,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143877270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"SAMBV: A fine-tuned SAM with interpolation consistency regularization for semi-supervised bi-ventricle segmentation from cardiac MRI","authors":"Yuhan Wang ,&nbsp;Shoujun Zhou ,&nbsp;Ke Lu ,&nbsp;Yuanquan Wang ,&nbsp;Lei Zhang ,&nbsp;Weipeng Liu ,&nbsp;Zhida Wang","doi":"10.1016/j.medengphy.2025.104341","DOIUrl":"10.1016/j.medengphy.2025.104341","url":null,"abstract":"<div><div>The SAM (segment anything model) is a foundation model for general purpose image segmentation, however, when it comes to a specific medical application, such as segmentation of both ventricles from the 2D cardiac MRI, the results are not satisfactory. The scarcity of labeled medical image data further increases the difficulty to apply the SAM to medical image processing. To address these challenges, we propose the SAMBV by fine-tuning the SAM for semi-supervised segmentation of bi-ventricle from the 2D cardiac MRI. The SAM is tuned in three aspects, (i) the position and feature adapters are introduced so that the SAM can adapt to bi-ventricle segmentation. (ii) a dual-branch encoder is incorporated to collect missing local feature information in SAM so as to improve bi-ventricle segmentation. (iii) the interpolation consistency regularization (ICR) semi-supervised manner is utilized, allowing the SAMBV to achieve competitive performance with only 40% of the labeled data in the ACDC dataset. Experimental results demonstrate that the proposed SAMBV achieves an average Dice score improvement of 17.6% over the original SAM, raising its performance from 74.49% to 92.09%. Furthermore, the SAMBV outperforms other supervised SAM fine-tuning methods, showing its effectiveness in semi-supervised medical image segmentation tasks. Notably, the proposed method is specifically designed for 2D MRI data.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104341"},"PeriodicalIF":1.7,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Functional brain response pattern under rubber hand illusion based on fNIRS","authors":"Pengju Liu ,&nbsp;Rong Liu ,&nbsp;Chih-Hsiu Cheng ,&nbsp;Lizhen Wang ,&nbsp;Yubo Fan","doi":"10.1016/j.medengphy.2025.104340","DOIUrl":"10.1016/j.medengphy.2025.104340","url":null,"abstract":"<div><div>The rubber hand illusion (RHI), where people experience a sense of ownership over a rubber hand, has been researched by various neuroimaging methods. Here we used functional near-infrared spectroscopy (fNIRS) to analyze the activation and functional connectivity of related brain regions under RHI. Meanwhile, three brain functional network parameters were calculated and analyzed: degree, clustering coefficient, and characteristic path length. fNIRS results showed that under RHI, the oxyhemoglobin (HbO) concentration increased in the prefrontal cortex (PFC), motor cortex (MC) and occipital lobe (OL). The functional connectivity between right PFC and bilateral OL was increased, while the connection level between left MC and bilateral OL was decreased. Brain network under RHI condition had smaller average degree, average clustering coefficient, and shorter average characteristic path length. Notably, the information processing and exchange functions of left MC seem to be weakened under RHI state, which was also partially corroborated by the reduced local efficiency shown in brain functional network analysis. Overall, we suggest that enhanced functional connectivity between the right MC, OL and PFC, and functional inhibition of the left MC were key to RHI production. The study significance lies in enhancing understanding of body ownership and sensory integration.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104340"},"PeriodicalIF":1.7,"publicationDate":"2025-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ensemble learning of deep CNN models and two stage level prediction of Cobb angle on surface topography in adolescents with idiopathic scoliosis","authors":"Mostafa Hassan ,&nbsp;Jose Maria Gonzalez Ruiz ,&nbsp;Nada Mohamed ,&nbsp;Thomaz Nogueira Burke ,&nbsp;Qipei Mei ,&nbsp;Lindsey Westover","doi":"10.1016/j.medengphy.2025.104332","DOIUrl":"10.1016/j.medengphy.2025.104332","url":null,"abstract":"<div><div>This study employs Convolutional Neural Networks (CNNs) as feature extractors with appended regression layers for the non-invasive prediction of Cobb Angle (CA) from Surface Topography (ST) scans in adolescents with Idiopathic Scoliosis (AIS). The aim is to minimize radiation exposure during critical growth periods by offering a reliable, non-invasive assessment tool. The efficacy of various CNN-based feature extractors—DenseNet121, EfficientNetB0, ResNet18, SqueezeNet, and a modified U-Net—was evaluated on a dataset of 654 ST scans using a regression analysis framework for accurate CA prediction. The dataset comprised 590 training and 64 testing scans. Performance was evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and accuracy in classifying scoliosis severity (mild, moderate, severe) based on CA measurements. The EfficientNetB0 feature extractor outperformed other models, demonstrating strong performance on the training set (<span><math><mtext>R</mtext><mo>=</mo><mn>0.96</mn></math></span>, R<span><math><mmultiscripts><mrow><mo>=</mo></mrow><mprescripts></mprescripts><none></none><mrow><mn>2</mn></mrow></mmultiscripts><mn>0.93</mn></math></span>) and achieving an MAE of <span><math><msup><mrow><mn>6.13</mn></mrow><mrow><mo>∘</mo></mrow></msup></math></span> and RMSE of <span><math><msup><mrow><mn>7.5</mn></mrow><mrow><mo>∘</mo></mrow></msup></math></span> on the test set. In terms of scoliosis severity classification, it achieved high precision (84.62%) and specificity (95.65% for mild cases and 82.98% for severe cases), highlighting its clinical applicability in AIS management. The regression-based approach using the EfficientNetB0 as a feature extractor presents a significant advancement for accurately determining CA from ST scans, offering a promising tool for improving scoliosis severity categorization and management in adolescents.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"140 ","pages":"Article 104332"},"PeriodicalIF":1.7,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143863885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A new network structure for Parkinson's handwriting image recognition","authors":"Xiao Jiang ,&nbsp;Haibin Yu ,&nbsp;Jiayu Yang ,&nbsp;Xiaoli Liu ,&nbsp;Zhu Li","doi":"10.1016/j.medengphy.2025.104333","DOIUrl":"10.1016/j.medengphy.2025.104333","url":null,"abstract":"<div><div>Parkinson's disease (PD) remains a condition without a cure, though its early manifestations can be managed effectively by medical professionals. This underscores the significance of early detection of PD. It has been widely demonstrated that handwriting analysis is a promising avenue for early PD diagnosis. In recent research, there has been a pivot towards leveraging artificial intelligence (AI) technologies for analyzing handwriting images to aid in diagnosing the disease. This study introduces an innovative network architecture specifically designed to capture the nuances of tremor and irregular spacing characteristic of PD patients' handwriting. By incorporating an attention mechanism, this network is capable of prioritizing different areas within the handwriting feature map, according to their diagnostic relevance. This approach significantly enhances the accuracy of detecting PD through handwriting analysis, with our model achieving an impressive mean accuracy rate of 96.5 %. When compared to traditional convolutional neural networks, our attention-based continuous convolutional network model demonstrates a substantial increase in diagnostic precision.</div></div>","PeriodicalId":49836,"journal":{"name":"Medical Engineering & Physics","volume":"139 ","pages":"Article 104333"},"PeriodicalIF":1.7,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143833284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}