Pub Date : 2025-03-24DOI: 10.1109/JTEHM.2025.3551783
{"title":"2024 Index IEEE Journal of Translational Engineering in Health and Medicine Vol. 12","authors":"","doi":"10.1109/JTEHM.2025.3551783","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3551783","url":null,"abstract":"","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"12 ","pages":"740-756"},"PeriodicalIF":3.7,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10937338","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143688092","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 : 2025-03-19DOI: 10.1109/JTEHM.2025.3571255
Luisa Neubig;Deirdre Larsen;Melda Kunduk;Andreas M. Kist
Objective: Dysphagia is a common and complex disorder that complicates both diagnoses and treatment. Consequently, the associated electronic health records (EHR) are often unstructured and complex, posing challenges for systematic data analysis.Methods and procedures: In this study, we employ natural language processing (NLP) techniques and large language models (LLMs) to automatically analyze clinical narratives and extract diagnostic information from a diverse set of EHRs. Our dataset includes medical records from 486 patients, representing a group with diverse dysphagic conditions. We analyze diagnoses provided in unstructured free text that do not follow a standardized structure. We utilize clustering algorithms on the extracted diagnostic features to identify distinct groups of patients who share similar pathophysiological swallowing dysfunctions.Results: We found that basic NLP techniques often provide limited insights due to the high variability of the data. In contrast, LLMs help to bridge the gap in understanding the nuanced medical information about dysphagia and related conditions. Although applying these advanced LLM models is not straightforward, our results demonstrate that leveraging closed-source models can effectively cluster different categories of dysphagia.Conclusion: Our study provides therefore evidence that LLMs are highly promising in future dysphagia research.Clinical impact: Dysphagia is a symptom associated with various diseases, though its underlying relationships remain unclear. This study demonstrates how analyzing large volumes of electronic health records can help clarify the causes of dysphagia and identify contributing factors. By applying natural language processing, we aim to enhance both understanding and treatment, supporting clinical staff in improving individualized care by identifying relevant patient cohorts. Clinical and Translational Impact Statement: This study uses LLMs to efficiently preprocess unstructured EHRs, improving dysphagia diagnosis and patient clustering. It aligns with Clinical Research, enhancing diagnostic speed and enabling personalized treatment.
{"title":"Unstructured Electronic Health Records of Dysphagic Patients Analyzed by Large Language Models","authors":"Luisa Neubig;Deirdre Larsen;Melda Kunduk;Andreas M. Kist","doi":"10.1109/JTEHM.2025.3571255","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3571255","url":null,"abstract":"Objective: Dysphagia is a common and complex disorder that complicates both diagnoses and treatment. Consequently, the associated electronic health records (EHR) are often unstructured and complex, posing challenges for systematic data analysis.Methods and procedures: In this study, we employ natural language processing (NLP) techniques and large language models (LLMs) to automatically analyze clinical narratives and extract diagnostic information from a diverse set of EHRs. Our dataset includes medical records from 486 patients, representing a group with diverse dysphagic conditions. We analyze diagnoses provided in unstructured free text that do not follow a standardized structure. We utilize clustering algorithms on the extracted diagnostic features to identify distinct groups of patients who share similar pathophysiological swallowing dysfunctions.Results: We found that basic NLP techniques often provide limited insights due to the high variability of the data. In contrast, LLMs help to bridge the gap in understanding the nuanced medical information about dysphagia and related conditions. Although applying these advanced LLM models is not straightforward, our results demonstrate that leveraging closed-source models can effectively cluster different categories of dysphagia.Conclusion: Our study provides therefore evidence that LLMs are highly promising in future dysphagia research.Clinical impact: Dysphagia is a symptom associated with various diseases, though its underlying relationships remain unclear. This study demonstrates how analyzing large volumes of electronic health records can help clarify the causes of dysphagia and identify contributing factors. By applying natural language processing, we aim to enhance both understanding and treatment, supporting clinical staff in improving individualized care by identifying relevant patient cohorts. Clinical and Translational Impact Statement: This study uses LLMs to efficiently preprocess unstructured EHRs, improving dysphagia diagnosis and patient clustering. It aligns with Clinical Research, enhancing diagnostic speed and enabling personalized treatment.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"237-245"},"PeriodicalIF":3.7,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11006689","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144171055","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 : 2025-03-15DOI: 10.1109/JTEHM.2025.3570611
Alex C. Barksdale;Natalie G. Ferris;Eli Mattingly;Monika Śliwiak;Bastien Guerin;Lawrence L. Wald;Mathias Davids;Valerie Klein
Magnetic fields switching at kilohertz frequencies induce electric fields in the body, which can cause peripheral nerve stimulation (PNS). Although magnetostimulation has been extensively studied below 10 kHz, the behavior of PNS at higher frequencies remains poorly understood. This study aims to investigate PNS thresholds at frequencies up to 88.1 kHz and to explore deviations from the widely accepted hyperbolic strength-duration curve (SDC).PNS thresholds were measured in the head of 8 human volunteers using a solenoidal coil at 16 distinct frequencies, ranging from 200 Hz to 88.1 kHz. A hyperbolic SDC was used as a reference to compare the frequency-dependent behavior of PNS thresholds.Contrary to the predictions of the hyperbolic SDC, PNS thresholds did not decrease monotonically with frequency. Instead, thresholds reached a minimum near 25 kHz, after which they increased by an average of 39% from 25 kHz to 88.1 kHz across subjects. This pattern indicates a significant deviation from previously observed behavior at lower frequencies.Our results suggest that PNS thresholds exhibit a non-monotonic frequency dependence at higher frequencies, diverging from the traditional hyperbolic SDC. These findings offer critical data for refining neurodynamic models and provide insights for setting PNS safety limits in applications like MRI gradient coils and magnetic particle imaging (MPI). Further investigation is needed to understand the biological mechanisms driving these deviations beyond 25 kHz.Clinical impact—These findings call for further basic research into biological mechanisms underlying high frequency PNS threshold trends, and supports refinement of safety guidelines for MRI and MPI systems for clinical implementation.
{"title":"Measurement of Peripheral Nerve Magnetostimulation Thresholds of a Head Solenoid Coil Between 200 Hz and 88.1 kHz","authors":"Alex C. Barksdale;Natalie G. Ferris;Eli Mattingly;Monika Śliwiak;Bastien Guerin;Lawrence L. Wald;Mathias Davids;Valerie Klein","doi":"10.1109/JTEHM.2025.3570611","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3570611","url":null,"abstract":"Magnetic fields switching at kilohertz frequencies induce electric fields in the body, which can cause peripheral nerve stimulation (PNS). Although magnetostimulation has been extensively studied below 10 kHz, the behavior of PNS at higher frequencies remains poorly understood. This study aims to investigate PNS thresholds at frequencies up to 88.1 kHz and to explore deviations from the widely accepted hyperbolic strength-duration curve (SDC).PNS thresholds were measured in the head of 8 human volunteers using a solenoidal coil at 16 distinct frequencies, ranging from 200 Hz to 88.1 kHz. A hyperbolic SDC was used as a reference to compare the frequency-dependent behavior of PNS thresholds.Contrary to the predictions of the hyperbolic SDC, PNS thresholds did not decrease monotonically with frequency. Instead, thresholds reached a minimum near 25 kHz, after which they increased by an average of 39% from 25 kHz to 88.1 kHz across subjects. This pattern indicates a significant deviation from previously observed behavior at lower frequencies.Our results suggest that PNS thresholds exhibit a non-monotonic frequency dependence at higher frequencies, diverging from the traditional hyperbolic SDC. These findings offer critical data for refining neurodynamic models and provide insights for setting PNS safety limits in applications like MRI gradient coils and magnetic particle imaging (MPI). Further investigation is needed to understand the biological mechanisms driving these deviations beyond 25 kHz.<italic><b>Clinical impact</b></i>—These findings call for further basic research into biological mechanisms underlying high frequency PNS threshold trends, and supports refinement of safety guidelines for MRI and MPI systems for clinical implementation.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"275-285"},"PeriodicalIF":3.7,"publicationDate":"2025-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11005621","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144299302","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 : 2025-03-12DOI: 10.1109/JTEHM.2025.3568922
Yuan-Jin Lin;Shih-Lun Chen;Yi-Cheng Mao;Tsung-Yi Chen;Cheng-Hao Peng;Tzu-Hsiang Tsai;Kuo-Chen Li;Chiung-An Chen;Wei-Chen Tu;Patricia Angela R. Abu
Extraction of the third molar of the mandible is one of the most common oral surgical procedures. Preoperative monitoring and assessment are crucial to mitigate neurological risks. Identifying whether the third molar in the mandible compresses the inferior alveolar nerve still relies on dental professionals, a task that is repetitive and time-consuming. Thus, the primary objective is to utilize dental panoramic radiography for image processing and classify whether the third molar compresses the inferior alveolar nerve, aiming to reduce the demand for CT images in symptom diagnosis and mitigate the risks associated with high-dose radiation. This study proposes an innovative dental panoramic radiography segmentation technique to locate the third molar position. Subsequently, an innovative edge masking enhancement method is used to extract features of the inferior alveolar nerve and the third molar. Moreover, a transformer-based image detection model to consider whether the third molar compresses the inferior alveolar nerve. The third molar position localization method achieved an accuracy rate of 97.92%, compared to recent research at least improved by 3.6% accuracy. Subsequently, innovative edge masking and image enhancement methods improve classification accuracy by 4.3%, when supplemented with computed tomography scan images for further evaluation, the maximum accuracy reached 98.45%, representing a 4.5% improvement compared to previous studies. The third molar position detection results will impact the identification of the inferior alveolar nerve compressed by the third molar. Through the innovative edge region segmentation algorithm can effectively distinguish this object, and the overall evaluation accuracy can be improved by approximately 3.8%.
{"title":"Precision Oral Medicine: A DPR Segmentation and Transfer Learning Approach for Detecting Third Molar Compress Inferior Alveolar Nerve","authors":"Yuan-Jin Lin;Shih-Lun Chen;Yi-Cheng Mao;Tsung-Yi Chen;Cheng-Hao Peng;Tzu-Hsiang Tsai;Kuo-Chen Li;Chiung-An Chen;Wei-Chen Tu;Patricia Angela R. Abu","doi":"10.1109/JTEHM.2025.3568922","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3568922","url":null,"abstract":"Extraction of the third molar of the mandible is one of the most common oral surgical procedures. Preoperative monitoring and assessment are crucial to mitigate neurological risks. Identifying whether the third molar in the mandible compresses the inferior alveolar nerve still relies on dental professionals, a task that is repetitive and time-consuming. Thus, the primary objective is to utilize dental panoramic radiography for image processing and classify whether the third molar compresses the inferior alveolar nerve, aiming to reduce the demand for CT images in symptom diagnosis and mitigate the risks associated with high-dose radiation. This study proposes an innovative dental panoramic radiography segmentation technique to locate the third molar position. Subsequently, an innovative edge masking enhancement method is used to extract features of the inferior alveolar nerve and the third molar. Moreover, a transformer-based image detection model to consider whether the third molar compresses the inferior alveolar nerve. The third molar position localization method achieved an accuracy rate of 97.92%, compared to recent research at least improved by 3.6% accuracy. Subsequently, innovative edge masking and image enhancement methods improve classification accuracy by 4.3%, when supplemented with computed tomography scan images for further evaluation, the maximum accuracy reached 98.45%, representing a 4.5% improvement compared to previous studies. The third molar position detection results will impact the identification of the inferior alveolar nerve compressed by the third molar. Through the innovative edge region segmentation algorithm can effectively distinguish this object, and the overall evaluation accuracy can be improved by approximately 3.8%.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"286-298"},"PeriodicalIF":3.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11000294","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144323021","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}
Digital health and AI-enabled technologies hold the promise of addressing gaps in healthcare, but balancing rapid market access with the need for safe, functional, and user-centered solutions remains a challenge [1], [2]. Regulatory requirements for device development and market approval demand detailed documentation and predetermined protocols, which can limit the adaptability developers require for iterative improvement and real-world testing with patients and healthcare professionals [1], [3], [4]—an approach that would be highly beneficial for digital and AI-enabled technologies. As a result, key factors like clinical workflow integration, interoperability, and usability with the real range of in-use devices are often overlooked or addressed in a cursory fashion [5].
{"title":"Letter to the Editor on “From Concept to Clinic: Living Labs and Regulatory Sandboxes for Health System Digitalization and the Integration of Innovative Devices Into Clinical Workflows”","authors":"Rebecca Mathias;Anett Schönfelder;Cindy Welzel;Stephen Gilbert","doi":"10.1109/JTEHM.2025.3557508","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3557508","url":null,"abstract":"Digital health and AI-enabled technologies hold the promise of addressing gaps in healthcare, but balancing rapid market access with the need for safe, functional, and user-centered solutions remains a challenge <xref>[1]</xref>, <xref>[2]</xref>. Regulatory requirements for device development and market approval demand detailed documentation and predetermined protocols, which can limit the adaptability developers require for iterative improvement and real-world testing with patients and healthcare professionals <xref>[1]</xref>, <xref>[3]</xref>, <xref>[4]</xref>—an approach that would be highly beneficial for digital and AI-enabled technologies. As a result, key factors like clinical workflow integration, interoperability, and usability with the real range of in-use devices are often overlooked or addressed in a cursory fashion <xref>[5]</xref>.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"214-215"},"PeriodicalIF":3.7,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10999161","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143937981","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 : 2025-03-07DOI: 10.1109/JTEHM.2025.3567578
Emma Reznick;T. Kevin Best;Robert D. Gregg
Objective: Configuring a prosthetic leg is an integral part of the fitting process, but the personalization of a multi-modal powered knee-ankle prosthesis is often too complex to realize in a clinical environment. This paper develops both the technical means to individualize a hybrid kinematic-impedance controller for variable-incline walking and sit-stand transitions, and an intuitive Clinical Tuning Interface (CTI) that allows prosthetists to directly modify the controller behavior. Methods and procedures: Utilizing an established method for predicting kinematic gait individuality alongside a new parallel approach for kinetic individuality, we personalize continuous-phase/task models of joint impedance (during stance) and kinematics (during swing) using tuned characteristics exclusively from level-ground walking. To take advantage of this method, we developed a CTI that translates common clinical tuning parameters into model adjustments for the walking and sit-stand controllers. We then conducted a case study where a prosthetist iteratively tuned the powered prosthesis to an above-knee amputee participant in a simulated clinical session involving sit-stand transitions and level walking, from which incline/decline walking features were automatically calibrated. Results: The prosthetist fully tuned the multi-activity prosthesis controller in under 20 min. Each iteration of tuning (i.e., observation, parameter adjustment, and model reprocessing) took 2 min on average for walking and 1 min on average for sit-stand. The tuned behavior changes were appropriately manifested in the commanded prosthesis torques, both at the manually tuned tasks and automatically tuned tasks (inclines). Conclusion: The CTI leveraged able-bodied trends to efficiently personalize a wide array of walking tasks and sit-stand transitions, demonstrating the efficiency necessary for powered knee-ankle prostheses to become clinically viable. Clinical impact: This paper introduces a clinical tuning interface that simplifies the tuning process for multimodal robotic prosthetic legs, reducing the time required from several hours to just 20 minutes thus improving clinical feasibility.
{"title":"A Clinical Tuning Framework for Continuous Kinematic and Impedance Control of a Powered Knee-Ankle Prosthesis","authors":"Emma Reznick;T. Kevin Best;Robert D. Gregg","doi":"10.1109/JTEHM.2025.3567578","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3567578","url":null,"abstract":"<bold>Objective:</b> Configuring a prosthetic leg is an integral part of the fitting process, but the personalization of a multi-modal powered knee-ankle prosthesis is often too complex to realize in a clinical environment. This paper develops both the technical means to individualize a hybrid kinematic-impedance controller for variable-incline walking and sit-stand transitions, and an intuitive Clinical Tuning Interface (CTI) that allows prosthetists to directly modify the controller behavior. <bold>Methods and procedures:</b> Utilizing an established method for predicting kinematic gait individuality alongside a new parallel approach for kinetic individuality, we personalize continuous-phase/task models of joint impedance (during stance) and kinematics (during swing) using tuned characteristics exclusively from level-ground walking. To take advantage of this method, we developed a CTI that translates common clinical tuning parameters into model adjustments for the walking and sit-stand controllers. We then conducted a case study where a prosthetist iteratively tuned the powered prosthesis to an above-knee amputee participant in a simulated clinical session involving sit-stand transitions and level walking, from which incline/decline walking features were automatically calibrated. <bold>Results:</b> The prosthetist fully tuned the multi-activity prosthesis controller in under 20 min. Each iteration of tuning (i.e., observation, parameter adjustment, and model reprocessing) took 2 min on average for walking and 1 min on average for sit-stand. The tuned behavior changes were appropriately manifested in the commanded prosthesis torques, both at the manually tuned tasks and automatically tuned tasks (inclines). <bold>Conclusion:</b> The CTI leveraged able-bodied trends to efficiently personalize a wide array of walking tasks and sit-stand transitions, demonstrating the efficiency necessary for powered knee-ankle prostheses to become clinically viable. <bold>Clinical impact:</b> This paper introduces a clinical tuning interface that simplifies the tuning process for multimodal robotic prosthetic legs, reducing the time required from several hours to just 20 minutes thus improving clinical feasibility.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"227-236"},"PeriodicalIF":3.7,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10990182","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144170997","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}
Continuous monitoring of fetal heart rate (FHR) and uterine contractions (UC), otherwise known as cardiotocography (CTG), is often used to assess the risk of fetal compromise during labor. However, interpreting CTG recordings visually is challenging for clinicians, given the complexity of CTG patterns, leading to poor sensitivity. Efforts to address this issue have focused on data-driven deep-learning methods to detect fetal compromise automatically. However, their progress is impeded by limited CTG training datasets and the absence of a standardized evaluation workflow, hindering algorithm comparisons. In this study, we use a private CTG dataset of 9,887 CTG recordings with pH measurements and 552 CTG recordings from the open-access CTU-UHB dataset to conduct a cross-database evaluation of six deep-learning models for fetal compromise detection. We explore the impact of input selection of FHR and UC signals, signal pre-processing, downsampling frequency, and the influence of removing intermediate pH samples from the training dataset. Our findings reveal that using only FHR and pre-processing FHR with artefact removal and interpolation provides a significant improvement to classification performance for some model architectures while excluding intermediate pH samples did not significantly improve performance for any model. From our comparison of the six models, ResNet exhibited the strongest fetal compromise classification performance across both databases at a downsampling rate of 1Hz. Finally, class activation maps from highly contributing signal regions in the ResNet model aligned with clinical knowledge of compromised FHR patterns, highlighting the model’s interpretability. These insights may serve as a standardized reference for developing and comparing future works in this domain. Clinical and Translational Impact: This study provides a standardized workflow for comparing deep-learning methods for CTG classification. Ensuring new methods show generalizability and interpretability will improve their robustness and applicability in clinical settings.
{"title":"Cross-Database Evaluation of Deep Learning Methods for Intrapartum Cardiotocography Classification","authors":"Lochana Mendis;Debjyoti Karmakar;Marimuthu Palaniswami;Fiona Brownfoot;Emerson Keenan","doi":"10.1109/JTEHM.2025.3548401","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3548401","url":null,"abstract":"Continuous monitoring of fetal heart rate (FHR) and uterine contractions (UC), otherwise known as cardiotocography (CTG), is often used to assess the risk of fetal compromise during labor. However, interpreting CTG recordings visually is challenging for clinicians, given the complexity of CTG patterns, leading to poor sensitivity. Efforts to address this issue have focused on data-driven deep-learning methods to detect fetal compromise automatically. However, their progress is impeded by limited CTG training datasets and the absence of a standardized evaluation workflow, hindering algorithm comparisons. In this study, we use a private CTG dataset of 9,887 CTG recordings with pH measurements and 552 CTG recordings from the open-access CTU-UHB dataset to conduct a cross-database evaluation of six deep-learning models for fetal compromise detection. We explore the impact of input selection of FHR and UC signals, signal pre-processing, downsampling frequency, and the influence of removing intermediate pH samples from the training dataset. Our findings reveal that using only FHR and pre-processing FHR with artefact removal and interpolation provides a significant improvement to classification performance for some model architectures while excluding intermediate pH samples did not significantly improve performance for any model. From our comparison of the six models, ResNet exhibited the strongest fetal compromise classification performance across both databases at a downsampling rate of 1Hz. Finally, class activation maps from highly contributing signal regions in the ResNet model aligned with clinical knowledge of compromised FHR patterns, highlighting the model’s interpretability. These insights may serve as a standardized reference for developing and comparing future works in this domain. Clinical and Translational Impact: This study provides a standardized workflow for comparing deep-learning methods for CTG classification. Ensuring new methods show generalizability and interpretability will improve their robustness and applicability in clinical settings.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"123-135"},"PeriodicalIF":3.7,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10912500","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143706796","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}
Objective: Investigating consistent methods and metrics for classifying Detrusor Overactivity (DO) events and developing an automated robust method for clinical measurements calculation from cystometry data in persons with spinal cord injury (SCI).Methods and procedures: A two-stage method for was proposed to detect DO events. In the first stage, DO peaks were detected using local linear models combined with thresholding criteria derived from clinical definitions and known artifacts. In the second stage, a segmentation method was proposed to detect the start and end time points of each DO event, marking the DO activity periods. As a result, complete clinical measurements, including bladder compliance, can be estimated automatically. The method was developed and tested on 77 anonymized urodynamic samples from SCI individuals (40 DO-positive, 37 DO-negative) with 158 annotated DO events.Results: On test data, in terms of the patient-level diagnosis of DO, the proposed method achieved an accuracy of 100%. Individual DO event detection achieved an average precision of 0.94 and recall of 0.72. Detrusor activity period identification showed a precision of 0.86 and a recall of 0.88. The task of automated bladder compliance estimation showed that the point-value-based method yields a lower median absolute error (MAE) compared to the proposed line-fitting-based method, with a MAE of 5.20 and 7.14 ml/cmH2O, respectively. Finally, for classifying bladder function into normal, low and severely low compliance, the proposed method had an accuracy of 88%.Conclusion: Our proposed local model fitting with thresholding based on clinical knowledge, achieved accurate automated results for cytometry data, which will enable objective assessment of routinely performed examinations.Clinical and Translational Impact Statement—This work proposes a fully automated detrusor overactivity diagnosis and feature extraction method. Empowering medical teams to consistently assess urodynamic studies while aiding disease characterization and enhancing clinical decision-making for SCI patients. Furthermore, it provides a mathematically defined method for extending the pipeline to other populations and standardizing clinical assessments.Category: Clinical Engineering, Medical Devices and Systems.
{"title":"Automated Evaluation of Urodynamic Examinations Through Local Linear Models: Validation on Spinal Cord Injury Individuals","authors":"Wensi Zhang;Jürgen Pannek;Jens Wöllner;Robert Riener;Diego Paez-Granados","doi":"10.1109/JTEHM.2025.3544486","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3544486","url":null,"abstract":"Objective: Investigating consistent methods and metrics for classifying Detrusor Overactivity (DO) events and developing an automated robust method for clinical measurements calculation from cystometry data in persons with spinal cord injury (SCI).Methods and procedures: A two-stage method for was proposed to detect DO events. In the first stage, DO peaks were detected using local linear models combined with thresholding criteria derived from clinical definitions and known artifacts. In the second stage, a segmentation method was proposed to detect the start and end time points of each DO event, marking the DO activity periods. As a result, complete clinical measurements, including bladder compliance, can be estimated automatically. The method was developed and tested on 77 anonymized urodynamic samples from SCI individuals (40 DO-positive, 37 DO-negative) with 158 annotated DO events.Results: On test data, in terms of the patient-level diagnosis of DO, the proposed method achieved an accuracy of 100%. Individual DO event detection achieved an average precision of 0.94 and recall of 0.72. Detrusor activity period identification showed a precision of 0.86 and a recall of 0.88. The task of automated bladder compliance estimation showed that the point-value-based method yields a lower median absolute error (MAE) compared to the proposed line-fitting-based method, with a MAE of 5.20 and 7.14 ml/cmH2O, respectively. Finally, for classifying bladder function into normal, low and severely low compliance, the proposed method had an accuracy of 88%.Conclusion: Our proposed local model fitting with thresholding based on clinical knowledge, achieved accurate automated results for cytometry data, which will enable objective assessment of routinely performed examinations.<bold><i>Clinical and Translational Impact Statement—</i></b>This work proposes a fully automated detrusor overactivity diagnosis and feature extraction method. Empowering medical teams to consistently assess urodynamic studies while aiding disease characterization and enhancing clinical decision-making for SCI patients. Furthermore, it provides a mathematically defined method for extending the pipeline to other populations and standardizing clinical assessments.Category: Clinical Engineering, Medical Devices and Systems.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"111-122"},"PeriodicalIF":3.7,"publicationDate":"2025-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10897996","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143688153","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}
Pilots face complex working environments during flight missions, which can easily lead to excessive workload and affect flight safety. Physiological signals are commonly used to evaluate a pilot’s workload because they are objective and can directly reflect physiological mental states. However, existing methods have shortcomings in temporal modeling, making it challenging to fully capture the dynamic characteristics of physiological signals over time. Moreover, fusing features of data from different modalities is also difficult.To address these problems, we proposed a temporal relation modeling and multimodal adversarial alignment network (TRM-MAAN) for pilot workload evaluation. Specifically, a Transformer-based temporal relationship modeling module was used to learn complex temporal relationships for better feature extraction. In addition, an adversarial alignment-based multi-modal fusion module was applied to capture and integrate multi-modal information, reducing distribution shifts between different modalities. The performance of the proposed TRM-MAAN method was evaluated via experiments of classifying three workload states using electroencephalogram (EEG) and electromyography (EMG) recordings of eight healthy pilots.Experimental results showed that the classification accuracy and F1 score of the proposed method were significantly better than the baseline model across different subjects, with an average recognition accuracy of $91.90~pm ~1.72%$ and an F1 score of $91.86~pm ~1.75%$ .This work provides essential technical support for improving the accuracy and robustness of pilot workload evaluation and introduces a promising way for enhancing flight safety, offering broad application prospects. Clinical and Translational Impact Statement: The proposed scheme provides a promising solution for workload evaluation based on electrophysiological signals, with potential applications in aiding the clinical monitoring of fatigue, mental status, cognitive psychology, and other disorders.
{"title":"Temporal Relation Modeling and Multimodal Adversarial Alignment Network for Pilot Workload Evaluation","authors":"Xinhui Li;Ao Li;Wenyu Fu;Xun Song;Fan Li;Qiang Ma;Yong Peng;Zhao LV","doi":"10.1109/JTEHM.2025.3542408","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3542408","url":null,"abstract":"Pilots face complex working environments during flight missions, which can easily lead to excessive workload and affect flight safety. Physiological signals are commonly used to evaluate a pilot’s workload because they are objective and can directly reflect physiological mental states. However, existing methods have shortcomings in temporal modeling, making it challenging to fully capture the dynamic characteristics of physiological signals over time. Moreover, fusing features of data from different modalities is also difficult.To address these problems, we proposed a temporal relation modeling and multimodal adversarial alignment network (TRM-MAAN) for pilot workload evaluation. Specifically, a Transformer-based temporal relationship modeling module was used to learn complex temporal relationships for better feature extraction. In addition, an adversarial alignment-based multi-modal fusion module was applied to capture and integrate multi-modal information, reducing distribution shifts between different modalities. The performance of the proposed TRM-MAAN method was evaluated via experiments of classifying three workload states using electroencephalogram (EEG) and electromyography (EMG) recordings of eight healthy pilots.Experimental results showed that the classification accuracy and F1 score of the proposed method were significantly better than the baseline model across different subjects, with an average recognition accuracy of <inline-formula> <tex-math>$91.90~pm ~1.72%$ </tex-math></inline-formula> and an F1 score of <inline-formula> <tex-math>$91.86~pm ~1.75%$ </tex-math></inline-formula>.This work provides essential technical support for improving the accuracy and robustness of pilot workload evaluation and introduces a promising way for enhancing flight safety, offering broad application prospects. Clinical and Translational Impact Statement: The proposed scheme provides a promising solution for workload evaluation based on electrophysiological signals, with potential applications in aiding the clinical monitoring of fatigue, mental status, cognitive psychology, and other disorders.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"85-97"},"PeriodicalIF":3.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10890988","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143553344","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}
Objective: Mild cognitive impairment (MCI) is characterized by early symptoms of attentional decline and may be distinguished through motor learning results. A relationship was reported between dexterous hand movements and cognitive function in older adults. Therefore, this study focuses on motor learning involving dexterous hand movements. As motor learning engages two distinct types of attention, external and internal, we aimed to develop an evaluation method that separates these attentional functions within motor learning. The objective of this study was to develop and verify the effectiveness of this evaluation method. The effectiveness was assessed by comparing two motor learning variables between a normal cognitive (NC) and MCI groups. Method: To evaluate motor learning through dexterous hand movements, we utilized the iWakka device. Two types of visual tracking tasks, repeat and random, were designed to evaluate motor learning from different aspects. The tracking errors in both tasks were quantitatively measured, and the initial and final improvement rates during motor learning were defined as the evaluation variables. The study included 28 MCI participants and 40 NC participants, and the effectiveness of the proposed method was verified by comparing results between the groups. Results: The repeat task revealed a significantly lower learning rate in MCI participants (p <0.01). In contrast, no significant difference was observed between MCI and NC participants in the random task (p =0.67). Conclusion: The evaluation method proposed in this study demonstrated the possibility of obtaining evaluation variables that indicate the characteristics of MCI. Clinical Impact: The methods proposed in this work are clinically relevant because the proposed evaluation system can make evaluation variables for discriminating cognitive decline in MCI. That it, the proposed approach can also be used to provide discrimination for cognitive decline in MCI.
{"title":"Quantification of Motor Learning in Hand Adjustability Movements: An Evaluation Variable for Discriminant Cognitive Decline","authors":"Kazuya Toshima;Yu Chokki;Toshiaki Wasaka;Tsukasa Tamaru;Yoshifumi Morita","doi":"10.1109/JTEHM.2025.3540203","DOIUrl":"https://doi.org/10.1109/JTEHM.2025.3540203","url":null,"abstract":"Objective: Mild cognitive impairment (MCI) is characterized by early symptoms of attentional decline and may be distinguished through motor learning results. A relationship was reported between dexterous hand movements and cognitive function in older adults. Therefore, this study focuses on motor learning involving dexterous hand movements. As motor learning engages two distinct types of attention, external and internal, we aimed to develop an evaluation method that separates these attentional functions within motor learning. The objective of this study was to develop and verify the effectiveness of this evaluation method. The effectiveness was assessed by comparing two motor learning variables between a normal cognitive (NC) and MCI groups. Method: To evaluate motor learning through dexterous hand movements, we utilized the iWakka device. Two types of visual tracking tasks, repeat and random, were designed to evaluate motor learning from different aspects. The tracking errors in both tasks were quantitatively measured, and the initial and final improvement rates during motor learning were defined as the evaluation variables. The study included 28 MCI participants and 40 NC participants, and the effectiveness of the proposed method was verified by comparing results between the groups. Results: The repeat task revealed a significantly lower learning rate in MCI participants (p <0.01). In contrast, no significant difference was observed between MCI and NC participants in the random task (p =0.67). Conclusion: The evaluation method proposed in this study demonstrated the possibility of obtaining evaluation variables that indicate the characteristics of MCI. Clinical Impact: The methods proposed in this work are clinically relevant because the proposed evaluation system can make evaluation variables for discriminating cognitive decline in MCI. That it, the proposed approach can also be used to provide discrimination for cognitive decline in MCI.","PeriodicalId":54255,"journal":{"name":"IEEE Journal of Translational Engineering in Health and Medicine-Jtehm","volume":"13 ","pages":"75-84"},"PeriodicalIF":3.7,"publicationDate":"2025-02-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10879071","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143465630","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：