Pub Date : 2025-06-04DOI: 10.1109/JTEHM.2025.3576596
Christopher Nielsen;Matthias Wilms;Nils D. Forkert
The retinal age gap (RAG; the difference between the retina’s biological and chronological age) has recently gained increased attention as a potential image-based, non-invasive, and accessible biomarker for a broad spectrum of ocular and non-ocular diseases. Traditionally, machine learning predictions of biological retinal age utilize convolutional neural network (CNN) architectures and data from color fundus photography (CFP). Despite being previously unexplored, the multimodal fusion of two-dimensional CFP with three-dimensional optical coherence tomography (OCT) data has significant potential to enhance retinal age prediction accuracy and the diagnostic utility of the RAG biomarker. Therefore, this work presents a novel foundation model-based framework for multimodal retinal age prediction. Technology or Method: Feature representations from CFP and OCT images were extracted using RETFound, a powerful foundation model for retinal image analysis. These representations were then combined using an innovative fusion strategy to train a lightweight linear regression head model for predicting retinal age. Training and evaluation of the developed multimodal retinal age prediction model was achieved using retinal images from over 80,000 participants in the UK Biobank. Results: The developed multimodal model sets a new benchmark in retinal age prediction (mean absolute error of 2.75 years), outperforming traditional CNN and single-modality approaches. Additionally, multimodal RAG values demonstrated superior performance in classifying patients with diabetes mellitus type 1, multiple sclerosis, and chronic kidney disease, highlighting the clinical relevance of the proposed multimodal approach for non-ocular disease detection. Conclusions: This work demonstrates that multimodal fusion of CFP and OCT significantly improves retinal age prediction and subsequent RAG-based analyses. By leveraging foundation models and multimodal retinal imaging, the proposed approach enhances disease classification accuracy and demonstrates the potential of integrating the RAG into clinical workflows as a scalable, non-invasive screening tool. Significance: The findings underscore the potential of multimodal retinal imaging to transform RAG into a clinically relevant and highly accessible biomarker for disease detection.
{"title":"A Novel Foundation Model-Based Framework for Multimodal Retinal Age Prediction","authors":"Christopher Nielsen;Matthias Wilms;Nils D. Forkert","doi":"10.1109/JTEHM.2025.3576596","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3576596","url":null,"abstract":"The retinal age gap (RAG; the difference between the retina’s biological and chronological age) has recently gained increased attention as a potential image-based, non-invasive, and accessible biomarker for a broad spectrum of ocular and non-ocular diseases. Traditionally, machine learning predictions of biological retinal age utilize convolutional neural network (CNN) architectures and data from color fundus photography (CFP). Despite being previously unexplored, the multimodal fusion of two-dimensional CFP with three-dimensional optical coherence tomography (OCT) data has significant potential to enhance retinal age prediction accuracy and the diagnostic utility of the RAG biomarker. Therefore, this work presents a novel foundation model-based framework for multimodal retinal age prediction. Technology or Method: Feature representations from CFP and OCT images were extracted using RETFound, a powerful foundation model for retinal image analysis. These representations were then combined using an innovative fusion strategy to train a lightweight linear regression head model for predicting retinal age. Training and evaluation of the developed multimodal retinal age prediction model was achieved using retinal images from over 80,000 participants in the UK Biobank. Results: The developed multimodal model sets a new benchmark in retinal age prediction (mean absolute error of 2.75 years), outperforming traditional CNN and single-modality approaches. Additionally, multimodal RAG values demonstrated superior performance in classifying patients with diabetes mellitus type 1, multiple sclerosis, and chronic kidney disease, highlighting the clinical relevance of the proposed multimodal approach for non-ocular disease detection. Conclusions: This work demonstrates that multimodal fusion of CFP and OCT significantly improves retinal age prediction and subsequent RAG-based analyses. By leveraging foundation models and multimodal retinal imaging, the proposed approach enhances disease classification accuracy and demonstrates the potential of integrating the RAG into clinical workflows as a scalable, non-invasive screening tool. Significance: The findings underscore the potential of multimodal retinal imaging to transform RAG into a clinically relevant and highly accessible biomarker for disease detection.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"299-309"},"PeriodicalIF":3.7,"publicationDate":"2025-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11023594","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144524371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-30DOI: 10.1109/JTEHM.2025.3565986
Mhairi Mcinnes;Dimitra Blana;Andrew Starkey;Edward K. Chadwick
Inertial sensors have the potential to be a useful clinical tool because they can facilitate human motion capture outside the research setting. A major barrier to the widespread application of inertial motion capture is the lack of accepted calibration methods for ensuring accuracy, in particular the lack of a common convention for calculating the rotational offset of the sensors, known as sensor-to-segment calibration. The purpose of this study was to develop and test a sensor-to-segment calibration method for upper limb motion capture which is practical for clinical applications.We developed a calibration method which depends mainly on the estimation of joint axes from arbitrary elbow motion, and partially on the design of custom attachment mounts to achieve physical alignment. With twenty healthy participants, we used OpenSim’s inertial sensor workflow to calculate joint kinematics, and evaluated the accuracy of the method through comparison with optical motion capture.We found the new calibration method resulted in upper limb kinematics with a median RMS error of 5–8°, and a median correlation coefficient of 0.977–0.987, which was significantly more accurate than a static pose calibration (p-value < 0.001).This work has demonstrated a method of calibration which is practical for clinical applications because it is quick to perform and does not depend on the subject’s ability to perform specific movements, or on the operator’s ability to carefully place sensors.Clinical Impact: The calibration method proposed in this work is a realistic option for the translation of inertial sensor technology into everyday clinical use.
{"title":"A Practical Sensor-to-Segment Calibration Method for Upper Limb Inertial Motion Capture in a Clinical Setting","authors":"Mhairi Mcinnes;Dimitra Blana;Andrew Starkey;Edward K. Chadwick","doi":"10.1109/JTEHM.2025.3565986","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3565986","url":null,"abstract":"Inertial sensors have the potential to be a useful clinical tool because they can facilitate human motion capture outside the research setting. A major barrier to the widespread application of inertial motion capture is the lack of accepted calibration methods for ensuring accuracy, in particular the lack of a common convention for calculating the rotational offset of the sensors, known as sensor-to-segment calibration. The purpose of this study was to develop and test a sensor-to-segment calibration method for upper limb motion capture which is practical for clinical applications.We developed a calibration method which depends mainly on the estimation of joint axes from arbitrary elbow motion, and partially on the design of custom attachment mounts to achieve physical alignment. With twenty healthy participants, we used OpenSim’s inertial sensor workflow to calculate joint kinematics, and evaluated the accuracy of the method through comparison with optical motion capture.We found the new calibration method resulted in upper limb kinematics with a median RMS error of 5–8°, and a median correlation coefficient of 0.977–0.987, which was significantly more accurate than a static pose calibration (p-value < 0.001).This work has demonstrated a method of calibration which is practical for clinical applications because it is quick to perform and does not depend on the subject’s ability to perform specific movements, or on the operator’s ability to carefully place sensors.Clinical Impact: The calibration method proposed in this work is a realistic option for the translation of inertial sensor technology into everyday clinical use.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"216-226"},"PeriodicalIF":3.7,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10981591","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143943979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Parkinson’s disease (PD) is characterized by gait disturbances with freezing of gait (FoG) being one of the most disabling symptoms. The FoG episode is often preceded by an increase in variability in Step Time. As the disease progresses, such gait impairment may become resistant to pharmacotherapy. Use of external cues is an alternative. Existing solutions deliver external cues in a continuous manner that might cause habituation effects, thereby emphasizing the need for on-demand cueing. Manual on-demand cueing upon freezing has been shown to be powerful in bringing an individual out of a freezing state. This can be achieved if one’s proneness to freeze before entering into freezing state can be sensed, and in-turn triggering an external cue on-demand. Motivated by this, we have developed a wearable device ($mathrm{SmartWalk}_{mathrm {VC}}$ ) that can sense such proneness based on variability in Step Time to offer a visual cue on-demand. We conducted a study involving 20 age-matched healthy individuals and those with PD who walked overground while wearing SmartWalkVC operated in three modes with regard to offering visual cue, namely (a) On-demand cueing, (b) Continuous cueing and (c) No cueing. The results of our study showed that with on-demand cueing, those with PD had minimum variability of Step Time among all the three modes unlike healthy individuals whose gait remained majorly unaffected by different cueing modes. Also, walking speed increased along with a reduction in FoG episodes for those with PD in the on-demand cueing mode compared with the other two modes.Clinical and Translational Impact Statement: Wearable SmartWalkVC quantifies one’s Step Time variability to offer visual cue on-demand, reducing one’s Freezing of Gait that can have clinical significance and be translated to impact one’s social presence.
{"title":"On-Demand Cueing Sensitive to Step Variability: Understanding Its Impact on Gait of Individuals With Parkinson’s Disease","authors":"Priya Pallavi;Ankita Raghuvanshi;Suhagiya Dharmik Kumar;Niravkumar Patel;Manasi Kanetkar;Rahul Chhatlani;Manish Rana;Sagar Betai;Roopa Rajan;Uttama Lahiri","doi":"10.1109/JTEHM.2025.3563381","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3563381","url":null,"abstract":"Parkinson’s disease (PD) is characterized by gait disturbances with freezing of gait (FoG) being one of the most disabling symptoms. The FoG episode is often preceded by an increase in variability in Step Time. As the disease progresses, such gait impairment may become resistant to pharmacotherapy. Use of external cues is an alternative. Existing solutions deliver external cues in a continuous manner that might cause habituation effects, thereby emphasizing the need for on-demand cueing. Manual on-demand cueing upon freezing has been shown to be powerful in bringing an individual out of a freezing state. This can be achieved if one’s proneness to freeze before entering into freezing state can be sensed, and in-turn triggering an external cue on-demand. Motivated by this, we have developed a wearable device (<inline-formula> <tex-math>$mathrm{SmartWalk}_{mathrm {VC}}$ </tex-math></inline-formula>) that can sense such proneness based on variability in Step Time to offer a visual cue on-demand. We conducted a study involving 20 age-matched healthy individuals and those with PD who walked overground while wearing SmartWalkVC operated in three modes with regard to offering visual cue, namely (a) On-demand cueing, (b) Continuous cueing and (c) No cueing. The results of our study showed that with on-demand cueing, those with PD had minimum variability of Step Time among all the three modes unlike healthy individuals whose gait remained majorly unaffected by different cueing modes. Also, walking speed increased along with a reduction in FoG episodes for those with PD in the on-demand cueing mode compared with the other two modes.Clinical and Translational Impact Statement: Wearable SmartWalkVC quantifies one’s Step Time variability to offer visual cue on-demand, reducing one’s Freezing of Gait that can have clinical significance and be translated to impact one’s social presence.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"183-192"},"PeriodicalIF":3.7,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10976342","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143896394","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-24DOI: 10.1109/JTEHM.2025.3563985
Sara Bernasconi;Giovanni Maria Oriolo;Giovanni Farina;Andrea Aliverti;Antonella Lomauro
Lymphedema, characterized by limb swelling, is typically treated with Complex Decongestive Therapy (CDT), which includes physical exercise. This study seeks to design and validate a wearable device aimed at enhancing CDT by monitoring patient adherence to prescribed exercises and tracking changes in the range of motion of the affected limbs. A wearable device, constituted by two boards with 2 IMUs, connected by a flexible flat cable, was designed and developed for placement across targeted joints. It communicates wirelessly with PCs, where raw data from IMUs are collected. Through the application of the Madgwick filter, orientation of the units is obtained and finally joints angles are computed. The device was validated through bench testing using an orthopedic goniometer and field testing with an optoelectronic system. The in vivo validation involved 18 volunteers, including 10 healthy individuals and 8 individuals with lymphedema, who performed flexion-extension movements and walked on a treadmill (at speeds of 3 km/h and 5 km/h). Bench testing demonstrated strong correlation and agreement (r2=0.999, mean percentage error = -0.51°, standard deviation = 2.00°). Once worn by the participants, the device enabled the measurement of joint angles during flexion-extension exercises (r2=0.852, mean percentage error = 1.44°, standard deviation = 11.7°) and the extraction of step counting, step time and toe off during walk at different speeds. The developed wearable device exhibited robust performance in both bench and field testing. This device, designed specifically for lymphedema patients, offers valuable insights into limb function and exercise adherence, potentially improving personalized treatment strategies.
{"title":"Design and Validation of a Wearable System for Enhanced Monitoring of Lower Limb Lymphedema","authors":"Sara Bernasconi;Giovanni Maria Oriolo;Giovanni Farina;Andrea Aliverti;Antonella Lomauro","doi":"10.1109/JTEHM.2025.3563985","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3563985","url":null,"abstract":"Lymphedema, characterized by limb swelling, is typically treated with Complex Decongestive Therapy (CDT), which includes physical exercise. This study seeks to design and validate a wearable device aimed at enhancing CDT by monitoring patient adherence to prescribed exercises and tracking changes in the range of motion of the affected limbs. A wearable device, constituted by two boards with 2 IMUs, connected by a flexible flat cable, was designed and developed for placement across targeted joints. It communicates wirelessly with PCs, where raw data from IMUs are collected. Through the application of the Madgwick filter, orientation of the units is obtained and finally joints angles are computed. The device was validated through bench testing using an orthopedic goniometer and field testing with an optoelectronic system. The in vivo validation involved 18 volunteers, including 10 healthy individuals and 8 individuals with lymphedema, who performed flexion-extension movements and walked on a treadmill (at speeds of 3 km/h and 5 km/h). Bench testing demonstrated strong correlation and agreement (r<sup>2</sup>=0.999, mean percentage error = -0.51°, standard deviation = 2.00°). Once worn by the participants, the device enabled the measurement of joint angles during flexion-extension exercises (r<sup>2</sup>=0.852, mean percentage error = 1.44°, standard deviation = 11.7°) and the extraction of step counting, step time and toe off during walk at different speeds. The developed wearable device exhibited robust performance in both bench and field testing. This device, designed specifically for lymphedema patients, offers valuable insights into limb function and exercise adherence, potentially improving personalized treatment strategies.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"193-201"},"PeriodicalIF":3.7,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10975766","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925066","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Accurate prediction of survival rates in esophageal cancer (EC) is crucial for guiding personalized treatment decisions. Deep learning-based survival models have gained increasing attention due to their powerful ability to capture complex embeddings in medical data. However, the primary limitation of current frameworks for predicting survival lies in their lack of attention to the contextual interactions between tumor and lymph node regions, which are vital for survival predictions. In the current study, we aimed to develop an effective EC survival risk prediction using only 3D computed tomography (CT) images.The proposed model consists of two essential components: 1) non-local feature aggregation module(NFAM) that integrates visual features from tumor and lymph nodes at both local and global scales, 2) graph-based spatial interaction module(GSIM) that explores the latent contextual interactions between tumors and lymph nodes.The experimental results demonstrate that our model achieves superior performance compared to state-of-the-art survival prediction methods, emphasizing its robust predictive capability. Moreover, we found that retaining lymph nodes with major axis $geq 8$ mm yields the best predictive results (C-index: 0.725), offering valuable guidance on choosing prognostic factors for esophageal cancer.For EC survival prediction using solely 3D CT images, integrating lymph node information with tumor information helps to improve the predictive performance of deep learning models.Clinical impact: The American Joint Committee on Cancer (TNM) classification serves as the primary framework for risk stratification, prognostic evaluation, and therapeutic decision-making in oncology. Nevertheless, this prognostic tool has demonstrated limited predictive accuracy in assessing long-term survival for esophageal carcinoma patients undergoing multimodal therapeutic regimens. Notably, even among those categorized within identical staging parameters, significant outcome heterogeneity persists, with survival trajectories diverging substantially across clinically matched populations. Our model serves as a complementary tool to the TNM staging system. By stratifying patients into distinct risk categories, this approach enables accurate prognosis assessment and provides critical guidance for postoperative adjuvant therapy decisions (such as whether to administer adjuvant radiotherapy or chemotherapy), thereby facilitating personalized treatment recommendations.
{"title":"Survival Prediction of Esophageal Cancer Using 3D CT Imaging: A Context-Aware Approach With Non-Local Feature Aggregation and Graph-Based Spatial Interaction","authors":"Fuce Guo;Chen Huang;Shengmei Lin;Yongmei Dai;Qianshun Chen;Shu Zhang;Xunyu XU","doi":"10.1109/JTEHM.2025.3562724","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3562724","url":null,"abstract":"Accurate prediction of survival rates in esophageal cancer (EC) is crucial for guiding personalized treatment decisions. Deep learning-based survival models have gained increasing attention due to their powerful ability to capture complex embeddings in medical data. However, the primary limitation of current frameworks for predicting survival lies in their lack of attention to the contextual interactions between tumor and lymph node regions, which are vital for survival predictions. In the current study, we aimed to develop an effective EC survival risk prediction using only 3D computed tomography (CT) images.The proposed model consists of two essential components: 1) non-local feature aggregation module(NFAM) that integrates visual features from tumor and lymph nodes at both local and global scales, 2) graph-based spatial interaction module(GSIM) that explores the latent contextual interactions between tumors and lymph nodes.The experimental results demonstrate that our model achieves superior performance compared to state-of-the-art survival prediction methods, emphasizing its robust predictive capability. Moreover, we found that retaining lymph nodes with major axis <inline-formula> <tex-math>$geq 8$ </tex-math></inline-formula>mm yields the best predictive results (C-index: 0.725), offering valuable guidance on choosing prognostic factors for esophageal cancer.For EC survival prediction using solely 3D CT images, integrating lymph node information with tumor information helps to improve the predictive performance of deep learning models.Clinical impact: The American Joint Committee on Cancer (TNM) classification serves as the primary framework for risk stratification, prognostic evaluation, and therapeutic decision-making in oncology. Nevertheless, this prognostic tool has demonstrated limited predictive accuracy in assessing long-term survival for esophageal carcinoma patients undergoing multimodal therapeutic regimens. Notably, even among those categorized within identical staging parameters, significant outcome heterogeneity persists, with survival trajectories diverging substantially across clinically matched populations. Our model serves as a complementary tool to the TNM staging system. By stratifying patients into distinct risk categories, this approach enables accurate prognosis assessment and provides critical guidance for postoperative adjuvant therapy decisions (such as whether to administer adjuvant radiotherapy or chemotherapy), thereby facilitating personalized treatment recommendations.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"202-213"},"PeriodicalIF":3.7,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10971407","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143913401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-15DOI: 10.1109/JTEHM.2025.3560877
Yang Li;Leo Yan Li-Han;Hua Tian
The clinical diagnosis of developmental dysplasia of the hip (DDH) typically involves manually measuring key radiological angles—Center-Edge (CE), Tönnis, and Sharp angles—from pelvic radiographs, a process that is time-consuming and susceptible to variability. This study aims to develop an automated system that integrates these measurements to enhance the accuracy and consistency of DDH diagnosis. We developed an end-to-end deep learning model for keypoint detection that accurately identifies eight anatomical keypoints from pelvic radiographs, enabling the automated calculation of CE, Tönnis, and Sharp angles. To support the diagnostic decision, we introduced a novel data-driven scoring system that combines the information from all three angles into a comprehensive and explainable diagnostic output. The system demonstrated superior consistency in angle measurements compared to a cohort of eight moderately experienced orthopedists. The intraclass correlation coefficients for the CE, Tönnis, and Sharp angles were 0.957 (95% CI: 0.952–0.962), 0.942 (95% CI: 0.937–0.947), and 0.966 (95% CI: 0.964–0.968), respectively. The system achieved a diagnostic F1 score of 0.863 (95% CI: 0.851–0.876), significantly outperforming the orthopedist group (0.777, 95% CI: 0.737–0.817, $p = 0.005$ ), as well as using clinical diagnostic criteria for each angle individually ($plt 0.001$ ). The proposed system provides reliable and consistent automated measurements of radiological angles and an explainable diagnostic output for DDH, outperforming moderately experienced clinicians.Clinical impact: This AI-powered solution reduces the variability and potential errors of manual measurements, offering clinicians a more consistent and interpretable tool for DDH diagnosis.
{"title":"Deep Learning-Based Automatic Diagnosis System for Developmental Dysplasia of the Hip","authors":"Yang Li;Leo Yan Li-Han;Hua Tian","doi":"10.1109/JTEHM.2025.3560877","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3560877","url":null,"abstract":"The clinical diagnosis of developmental dysplasia of the hip (DDH) typically involves manually measuring key radiological angles—Center-Edge (CE), Tönnis, and Sharp angles—from pelvic radiographs, a process that is time-consuming and susceptible to variability. This study aims to develop an automated system that integrates these measurements to enhance the accuracy and consistency of DDH diagnosis. We developed an end-to-end deep learning model for keypoint detection that accurately identifies eight anatomical keypoints from pelvic radiographs, enabling the automated calculation of CE, Tönnis, and Sharp angles. To support the diagnostic decision, we introduced a novel data-driven scoring system that combines the information from all three angles into a comprehensive and explainable diagnostic output. The system demonstrated superior consistency in angle measurements compared to a cohort of eight moderately experienced orthopedists. The intraclass correlation coefficients for the CE, Tönnis, and Sharp angles were 0.957 (95% CI: 0.952–0.962), 0.942 (95% CI: 0.937–0.947), and 0.966 (95% CI: 0.964–0.968), respectively. The system achieved a diagnostic F1 score of 0.863 (95% CI: 0.851–0.876), significantly outperforming the orthopedist group (0.777, 95% CI: 0.737–0.817, <inline-formula> <tex-math>$p = 0.005$ </tex-math></inline-formula>), as well as using clinical diagnostic criteria for each angle individually (<inline-formula> <tex-math>$plt 0.001$ </tex-math></inline-formula>). The proposed system provides reliable and consistent automated measurements of radiological angles and an explainable diagnostic output for DDH, outperforming moderately experienced clinicians.Clinical impact: This AI-powered solution reduces the variability and potential errors of manual measurements, offering clinicians a more consistent and interpretable tool for DDH diagnosis.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"174-182"},"PeriodicalIF":3.7,"publicationDate":"2025-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10965781","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143896492","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-10DOI: 10.1109/JTEHM.2025.3559693
Felipe Perez;Jorge Morisaki;Haitham Kanakri;Maher Rizkalla;Ahmed Abdalla
Late Onset Alzheimer’s Disease (LOAD) is the most common cause of dementia, characterized by the deposition of plaques primarily of neurotoxic amyloid-$beta $ ($Abeta $ ) peptide and tau protein. Our objective is to develop a noninvasive therapy to decrease the toxic A$beta $ levels, using repeated electromagnetic field stimulation (REMFS) in the brain of Alzheimer’s disease patients. We previously examined the effects of REMFS on $Abeta $ levels in primary human brain (PHB) cultures at different frequencies, powers, and specific absorption rates (SAR). PHB cultures at day in vitro (DIV7) treated with 64 MHz with a SAR of 0.6 W/Kg, one hour daily for 14 days (DIV 21) had significantly reduced (p =0.001) levels of secreted $Abeta $ -42 and $Abeta $ -40 peptide without evidence of toxicity. The EMF frequency and power, and SAR levels used in our work is utilized in MRI’s, thus suggesting REMFS can be further developed in clinical settings to lower ($Abeta $ ) levels and improve the memory in AD patients. These findings and numerous studies in rodent AD models prompted us to design a portable RF device, appropriate for human use, that will deliver a homogeneous RF power deposition with a SAR value of 0.4-0.9 W/kg to all human brain memory areas, lower ($Abeta $ ) levels, and potentially improve memory in human AD patients.The research took place at the Indiana University School of Medicine (IUSM) and Purdue University Indianapolis. The first phase was done in PHB cultures at the IUSM. Through this phase, we found that a 64 MHz frequency and an RF power deposition with a SAR of 0.4-0.6 W/kg reduced the (A$beta $ ) levels potentially impacting Alzheimer’s disease. The second phase of the project was conducted at Purdue University, we used ANSYS HFSS (High Frequency Simulation System) to design the devices that produced an appropriate penetration depth, polarization, and power deposition with a SAR of 0.4-0.9 W/kg to all memory brain areas of several numerical models. In Phase II-B will validate the device in a physical phantom. Phase III will require the FDA approval and application in clinical trials.The research parameters were translated into a designed product that fits comfortably in human head and fed from an external RF source that generates an RF power deposition with a SAR of 0.4-0.9 W/kg to a realistic numerical brain. The engineering design is flexible by varying the leg capacitors of the Meander Line Antenna (MLA) devices. Thermal outcomes of the resu
{"title":"A Novel Design of a Portable Birdcage via Meander Line Antenna (MLA) to Lower Beta Amyloid (Aβ) in Alzheimer’s Disease","authors":"Felipe Perez;Jorge Morisaki;Haitham Kanakri;Maher Rizkalla;Ahmed Abdalla","doi":"10.1109/JTEHM.2025.3559693","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3559693","url":null,"abstract":"Late Onset Alzheimer’s Disease (LOAD) is the most common cause of dementia, characterized by the deposition of plaques primarily of neurotoxic amyloid-<inline-formula> <tex-math>$beta $ </tex-math></inline-formula> (<inline-formula> <tex-math>$Abeta $ </tex-math></inline-formula>) peptide and tau protein. Our objective is to develop a noninvasive therapy to decrease the toxic A<inline-formula> <tex-math>$beta $ </tex-math></inline-formula> levels, using repeated electromagnetic field stimulation (REMFS) in the brain of Alzheimer’s disease patients. We previously examined the effects of REMFS on <inline-formula> <tex-math>$Abeta $ </tex-math></inline-formula> levels in primary human brain (PHB) cultures at different frequencies, powers, and specific absorption rates (SAR). PHB cultures at day in vitro (DIV7) treated with 64 MHz with a SAR of 0.6 W/Kg, one hour daily for 14 days (DIV 21) had significantly reduced (p =0.001) levels of secreted <inline-formula> <tex-math>$Abeta $ </tex-math></inline-formula>-42 and <inline-formula> <tex-math>$Abeta $ </tex-math></inline-formula>-40 peptide without evidence of toxicity. The EMF frequency and power, and SAR levels used in our work is utilized in MRI’s, thus suggesting REMFS can be further developed in clinical settings to lower (<inline-formula> <tex-math>$Abeta $ </tex-math></inline-formula>) levels and improve the memory in AD patients. These findings and numerous studies in rodent AD models prompted us to design a portable RF device, appropriate for human use, that will deliver a homogeneous RF power deposition with a SAR value of 0.4-0.9 W/kg to all human brain memory areas, lower (<inline-formula> <tex-math>$Abeta $ </tex-math></inline-formula>) levels, and potentially improve memory in human AD patients.The research took place at the Indiana University School of Medicine (IUSM) and Purdue University Indianapolis. The first phase was done in PHB cultures at the IUSM. Through this phase, we found that a 64 MHz frequency and an RF power deposition with a SAR of 0.4-0.6 W/kg reduced the (A<inline-formula> <tex-math>$beta $ </tex-math></inline-formula>) levels potentially impacting Alzheimer’s disease. The second phase of the project was conducted at Purdue University, we used ANSYS HFSS (High Frequency Simulation System) to design the devices that produced an appropriate penetration depth, polarization, and power deposition with a SAR of 0.4-0.9 W/kg to all memory brain areas of several numerical models. In Phase II-B will validate the device in a physical phantom. Phase III will require the FDA approval and application in clinical trials.The research parameters were translated into a designed product that fits comfortably in human head and fed from an external RF source that generates an RF power deposition with a SAR of 0.4-0.9 W/kg to a realistic numerical brain. The engineering design is flexible by varying the leg capacitors of the Meander Line Antenna (MLA) devices. Thermal outcomes of the resu","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"158-173"},"PeriodicalIF":3.7,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10962220","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143883508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-02DOI: 10.1109/JTEHM.2025.3557250
Giuseppe Turini;Marina Carbone;Sara Condino;Donato Gallone;Vincenzo Ferrari;Marco Gesi;Michelangelo Scaglione;Paolo Parchi;Rosanna Maria Viglialoro
Objective: Motivation and adherence are crucial for effective rehabilitation, yet engagement remains a challenge in upper limb physiotherapy. Serious Games (SGs) have emerged as a promising tool to enhance patient motivation. This study evaluates Painting Discovery, a projected augmented reality (AR) SG for shoulder rehabilitation, assessing engagement, ergonomics, and its potential to differentiate motor performance between healthy and those with rheumatoid arthritis, bursitis, subacromial impingement, rotator cuff tear, or calcific tendinopathy. Additionally, it examines improvements in pathological subjects following physiotherapy. Method: Sixteen healthy and seven pathological subjects participated. Engagement, ergonomics, and satisfaction were assessed using Likert-scale questionnaires. Motor performance was evaluated through completion time, speed, acceleration, and normalized jerk. Four pathological subjects underwent pre- and post-physiotherapy assessments over six weeks. Results: SG was highly engaging and ergonomic, with no significant differences based on prior video game or AR experience. The pathological group had longer completion times ($56.49~pm ~37.85$ s vs. $39.02~pm ~24.21$ s, p < 0.001), lower acceleration ($1.11~pm ~0.92$ m/s2 vs. $0.79~pm ~0.56$ m/s2, p < 0.001), and higher jerk ($6.68times 107~pm ~1.37times 108$ m/s3 vs. $9.22times 106~pm ~2.51times 107$ m/s3, p = 0.025) then healthy subjects. After physiotherapy, completion time and normalized jerk indicated enhanced efficiency and control. Conclusions: Painting Discovery shows strong potential as an engaging, accessible rehabilitation tool. While effective in differentiating motor impairments, its small sample size and horizontal-plane movement focus limit broader conclusions. Future studies should expand participation, incorporate vertical-plane movements, and refine performance metrics for clinical validation.
目的：动机和坚持是有效康复的关键，但参与上肢物理治疗仍然是一个挑战。严肃游戏（Serious Games, SGs）已成为增强患者动机的一种有前景的工具。本研究评估了用于肩部康复的增强现实（AR） SG - Painting Discovery，评估了参与性、人体工程学及其区分健康人与类风湿关节炎、滑囊炎、肩胛下撞击、肩袖撕裂或钙化肌腱病患者运动表现的潜力。此外，它还检查了物理治疗后病理受试者的改善。方法：健康受试者16例，病理受试者7例。参与、人体工程学和满意度采用李克特量表问卷进行评估。运动性能通过完成时间、速度、加速度和标准抽动来评估。四名病理受试者在六周内接受了物理治疗前后的评估。结果：SG是高度参与和符合人体工程学，没有显著差异基于先前的视频游戏或AR经验。病理组完成时间较健康组长（56.49~ 37.85$ s vs. 39.02~ 24.21$ s, p < 0.001），加速度较低（1.11~pm ~0.92$ m/s2 vs. 0.79~pm ~0.56$ m/s2, p < 0.001），跳速较高（6.68 × 107~pm ~1.37 × 108$ m/s3 vs. 9.22 × 106~pm ~2.51 × 107$ m/s3, p = 0.025）。物理治疗后，完成时间和正常抽搐表明效率和控制力增强。结论：绘画发现显示出强大的潜力，作为一个有吸引力的，可访问的康复工具。虽然在区分运动障碍方面是有效的，但它的小样本量和水平平面运动焦点限制了更广泛的结论。未来的研究应扩大参与，纳入垂直平面运动，并完善临床验证的性能指标。
{"title":"Projected AR Serious Game “Painting Discovery” for Shoulder Rehabilitation: Assessment With Technicians, Physiotherapists, and Patients","authors":"Giuseppe Turini;Marina Carbone;Sara Condino;Donato Gallone;Vincenzo Ferrari;Marco Gesi;Michelangelo Scaglione;Paolo Parchi;Rosanna Maria Viglialoro","doi":"10.1109/JTEHM.2025.3557250","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3557250","url":null,"abstract":"Objective: Motivation and adherence are crucial for effective rehabilitation, yet engagement remains a challenge in upper limb physiotherapy. Serious Games (SGs) have emerged as a promising tool to enhance patient motivation. This study evaluates Painting Discovery, a projected augmented reality (AR) SG for shoulder rehabilitation, assessing engagement, ergonomics, and its potential to differentiate motor performance between healthy and those with rheumatoid arthritis, bursitis, subacromial impingement, rotator cuff tear, or calcific tendinopathy. Additionally, it examines improvements in pathological subjects following physiotherapy. Method: Sixteen healthy and seven pathological subjects participated. Engagement, ergonomics, and satisfaction were assessed using Likert-scale questionnaires. Motor performance was evaluated through completion time, speed, acceleration, and normalized jerk. Four pathological subjects underwent pre- and post-physiotherapy assessments over six weeks. Results: SG was highly engaging and ergonomic, with no significant differences based on prior video game or AR experience. The pathological group had longer completion times (<inline-formula> <tex-math>$56.49~pm ~37.85$ </tex-math></inline-formula>s vs. <inline-formula> <tex-math>$39.02~pm ~24.21$ </tex-math></inline-formula>s, p < 0.001), lower acceleration (<inline-formula> <tex-math>$1.11~pm ~0.92$ </tex-math></inline-formula> m/s2 vs. <inline-formula> <tex-math>$0.79~pm ~0.56$ </tex-math></inline-formula> m/s2, p < 0.001), and higher jerk (<inline-formula> <tex-math>$6.68times 107~pm ~1.37times 108$ </tex-math></inline-formula> m/s3 vs. <inline-formula> <tex-math>$9.22times 106~pm ~2.51times 107$ </tex-math></inline-formula> m/s3, p = 0.025) then healthy subjects. After physiotherapy, completion time and normalized jerk indicated enhanced efficiency and control. Conclusions: Painting Discovery shows strong potential as an engaging, accessible rehabilitation tool. While effective in differentiating motor impairments, its small sample size and horizontal-plane movement focus limit broader conclusions. Future studies should expand participation, incorporate vertical-plane movements, and refine performance metrics for clinical validation.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"149-157"},"PeriodicalIF":3.7,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10947717","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-03-28DOI: 10.1109/JTEHM.2025.3574553
Jack Curley;Esteban Gomez;Laith Adnan;Isabelle Ablao;Jayden Sumbillo;Henry York;Hakan Töreyin
Objective: This study evaluates the feasibility of a noninvasive system for monitoring diaphragmatic efficiency in people with cervical spinal cord injury (CSCI). Methods: Two versions of a portable hardware system were developed using impedance pneumography (IP) to measure tidal volume (TV) and surface electromyography (sEMG) to assess diaphragm electrical activity (EAdi). Version 1 was used to determine optimal electrode positions, while Version 2 integrated these sensor systems into a compact, portable design. Data from eight healthy male participants were analyzed to assess the correlation and accuracy of TV and respiration rate (RR) prediction using IP and the correlation between sEMG signals and maximum inspiratory pressure (MIP). Results: For IP, measurements between the upper sternum and the midclavicular line (MCL) at the 4th intercostal (IC) space showed the highest correlation with true tidal volume. For sEMG, measurements between the mid-sternum and the 6th IC space demonstrated the strongest correlation with MIP. The integrated version 2 hardware demonstrates simultaneous IP and sEMG measurement while dissipating 2.17 mW. Discussion/Conclusion: The proposed system and the results presented may lead to a practical, cost-effective solution for continuous diaphragmatic efficiency monitoring, and thus enabling home-based respiratory care of CSCI patients. Clinical and Translational Impact Statement– This work presents the feasibility of building a wearable system that can unobtrusively monitor diaphragmatic efficiency, and thus enabling noninvasive, cost-effective, and home-based respiratory care for CSCI patients, facilitating early intervention and improved long-term health outcomes. This study is categorized under the early/pre-clinical research category of the NIH Clinical spectrum.
{"title":"Feasibility Analysis of a Portable Diaphragmatic Efficiency Monitor for CSCI Patients","authors":"Jack Curley;Esteban Gomez;Laith Adnan;Isabelle Ablao;Jayden Sumbillo;Henry York;Hakan Töreyin","doi":"10.1109/JTEHM.2025.3574553","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3574553","url":null,"abstract":"Objective: This study evaluates the feasibility of a noninvasive system for monitoring diaphragmatic efficiency in people with cervical spinal cord injury (CSCI). Methods: Two versions of a portable hardware system were developed using impedance pneumography (IP) to measure tidal volume (TV) and surface electromyography (sEMG) to assess diaphragm electrical activity (EAdi). Version 1 was used to determine optimal electrode positions, while Version 2 integrated these sensor systems into a compact, portable design. Data from eight healthy male participants were analyzed to assess the correlation and accuracy of TV and respiration rate (RR) prediction using IP and the correlation between sEMG signals and maximum inspiratory pressure (MIP). Results: For IP, measurements between the upper sternum and the midclavicular line (MCL) at the 4th intercostal (IC) space showed the highest correlation with true tidal volume. For sEMG, measurements between the mid-sternum and the 6th IC space demonstrated the strongest correlation with MIP. The integrated version 2 hardware demonstrates simultaneous IP and sEMG measurement while dissipating 2.17 mW. Discussion/Conclusion: The proposed system and the results presented may lead to a practical, cost-effective solution for continuous diaphragmatic efficiency monitoring, and thus enabling home-based respiratory care of CSCI patients. Clinical and Translational Impact Statement– This work presents the feasibility of building a wearable system that can unobtrusively monitor diaphragmatic efficiency, and thus enabling noninvasive, cost-effective, and home-based respiratory care for CSCI patients, facilitating early intervention and improved long-term health outcomes. This study is categorized under the early/pre-clinical research category of the NIH Clinical spectrum.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"246-250"},"PeriodicalIF":3.7,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11017366","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144272970","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Physical pain, particularly musculoskeletal pain, negatively impacts the activities of daily life and quality of life of elderly people. Because pain is a subjective sensation and there are no standard assessment procedures to detect pain, we attempted to quantitatively determine the actual state of chronic pain caused by musculoskeletal organs and related factors based on questionnaires. First, we studied techniques for diagnosing diseases by monitoring the involuntary characteristics of the voice. Then, we applied the technique based on voice characteristics and proposed a voice index to detect chronic musculoskeletal pain. The voice index was derived based on the assumption that physiological changes due to chronic musculoskeletal pain also affect the vocal cords. Subjects in this study were adults, 65 years of age or older, with chronic pain in the musculoskeletal system (lumbar and/or knees). A large-scale population-based cohort study was conducted in 2019. Voice characteristics were extracted from the recorded voices of the subjects, and the characteristics with similar properties were organized into several principal components using principal component analysis. The principal components were further combined using logistic regression analysis to propose a voice index that discriminates between normal subjects and subjects suffering from chronic musculoskeletal pain. A discrimination accuracy of approximately 80% was obtained using the dataset corresponding to the participants with knee pain only, and a discrimination accuracy of approximately 70% was obtained during cross-validation of the same dataset. The proposed voice index may serve as a novel tool for detecting chronic musculoskeletal pain. Clinical impact: The voice-based pain detection holds clinical significance owing to its noninvasive nature, ease of administration, and potential to efficiently assess large populations within a short time frame.
{"title":"Detection of Chronic Musculoskeletal Pain Using Voice Characteristics","authors":"Masakazu Higuchi;Toshiko Iidaka;Chiaki Horii;Gaku Tanegashima;Hiroyuki Oka;Hiroshi Hashizume;Hiroshi Yamada;Munehito Yoshida;Sakae Tanaka;Noriko Yoshimura;Mitsuteru Nakamura;Shinichi Tokuno","doi":"10.1109/JTEHM.2025.3553892","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3553892","url":null,"abstract":"Physical pain, particularly musculoskeletal pain, negatively impacts the activities of daily life and quality of life of elderly people. Because pain is a subjective sensation and there are no standard assessment procedures to detect pain, we attempted to quantitatively determine the actual state of chronic pain caused by musculoskeletal organs and related factors based on questionnaires. First, we studied techniques for diagnosing diseases by monitoring the involuntary characteristics of the voice. Then, we applied the technique based on voice characteristics and proposed a voice index to detect chronic musculoskeletal pain. The voice index was derived based on the assumption that physiological changes due to chronic musculoskeletal pain also affect the vocal cords. Subjects in this study were adults, 65 years of age or older, with chronic pain in the musculoskeletal system (lumbar and/or knees). A large-scale population-based cohort study was conducted in 2019. Voice characteristics were extracted from the recorded voices of the subjects, and the characteristics with similar properties were organized into several principal components using principal component analysis. The principal components were further combined using logistic regression analysis to propose a voice index that discriminates between normal subjects and subjects suffering from chronic musculoskeletal pain. A discrimination accuracy of approximately 80% was obtained using the dataset corresponding to the participants with knee pain only, and a discrimination accuracy of approximately 70% was obtained during cross-validation of the same dataset. The proposed voice index may serve as a novel tool for detecting chronic musculoskeletal pain. Clinical impact: The voice-based pain detection holds clinical significance owing to its noninvasive nature, ease of administration, and potential to efficiently assess large populations within a short time frame.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"136-148"},"PeriodicalIF":3.7,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10937750","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143801054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：