Shovito Barua Soumma
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shovito Barua Soumma has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Hassan Ghasemzadeh
Arizona State University
6 shared publications - Abdullah Mamun
Arizona State University, USA
3 shared publications Hassan Ghasemzadeh
College of Health Solutions, Arizona State University, Phoenix, AZ, USA
3 shared publications- Asiful Arefeen
College of Health Solutions, Arizona State University, Phoenix, AZ, USA
2 shared publications - Daniel S. Peterson
College of Health Solutions, Arizona State University, Phoenix, AZ, USA
2 shared publications - Ebrahim Farahmand
College of Health Solutions, Arizona State University, Phoenix, AZ, USA
2 shared publications
Latest Publications
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (9) , 14737-14753
Hybrid Attention Model Using Feature Decomposition and Knowledge Distillation for Blood Glucose Forecasting
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (9) , 14737-14753
Hybrid Attention Model Using Feature Decomposition and Knowledge Distillation for Blood Glucose Forecasting
- Open Access
- 28 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Engineering in Medicine and Biology
- Vol. PP, 1-10
Counterfactual Modeling with Fine-Tuned LLMs for Health Intervention Design and Sensor Data Augmentation
Counterfactual explanations (CFEs) provide human-centric interpretability by identifying the minimal, actionable changes required to alter a machine learning model's prediction. Therefore, CFs can be used as (i) interventions for abnormality prevention and (ii) augmented data for training robust models. we conduct a comprehensive evaluation of CF generation using large language models (LLMs), including GPT-4 (zero-shot and few-shot) and two open-source models-BioMistral-7B and LLaMA-3.1-8B-in both pretrained and fine-tuned configurations. Using the multimodal AI-READI clinical dataset, we assess CFs across three dimensions: intervention quality, feature diversity, and augmentation effectiveness. Fine-tuned LLMs, particularly LLaMA-3.1-8B, produce CFs with high plausibility (up to 99%), strong validity (up to 0.99), and realistic, behaviorally modifiable feature adjustments. When used for data augmentation under controlled label-scarcity settings, LLM-generated CFs substantially restore classifier performance, yielding an average 20% F1 recovery across three scarcity scenarios. Compared with optimization-based baselines such as DiCE, CFNOW, and NICE, LLMs offer a flexible, model-agnostic approach that generates more clinically plausible, behaviorally actionable and semantically coherent counterfactuals. Overall, this work demonstrates the promise of LLM-driven counterfactuals for both interpretable intervention design and data-efficient model training in sensor-based digital health.
- 28 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Engineering in Medicine and Biology
- Vol. PP, 1-10
Open AccessCounterfactual Modeling with Fine-Tuned LLMs for Health Intervention Design and Sensor Data Augmentation
Counterfactual explanations (CFEs) provide human-centric interpretability by identifying the minimal, actionable changes required to alter a machine learning model's prediction. Therefore, CFs can be used as (i) interventions for abnormality prevention and (ii) augmented data for training robust models. we conduct a comprehensive evaluation of CF generation using large language models (LLMs), including GPT-4 (zero-shot and few-shot) and two open-source models-BioMistral-7B and LLaMA-3.1-8B-in both pretrained and fine-tuned configurations. Using the multimodal AI-READI clinical dataset, we assess CFs across three dimensions: intervention quality, feature diversity, and augmentation effectiveness. Fine-tuned LLMs, particularly LLaMA-3.1-8B, produce CFs with high plausibility (up to 99%), strong validity (up to 0.99), and realistic, behaviorally modifiable feature adjustments. When used for data augmentation under controlled label-scarcity settings, LLM-generated CFs substantially restore classifier performance, yielding an average 20% F1 recovery across three scarcity scenarios. Compared with optimization-based baselines such as DiCE, CFNOW, and NICE, LLMs offer a flexible, model-agnostic approach that generates more clinically plausible, behaviorally actionable and semantically coherent counterfactuals. Overall, this work demonstrates the promise of LLM-driven counterfactuals for both interpretable intervention design and data-efficient model training in sensor-based digital health.
- Open Access
- 16 May 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Computing for Healthcare
- Vol. 7 (3) , 1-24
Self-Supervised Learning and Opportunistic Inference for Continuous Monitoring of Freezing of Gait in Parkinson’s Disease
Parkinson’s disease (PD) significantly affects patients’ quality of life through debilitating motor symptoms, such as Freezing of Gait (FoG). Continuous, in-home monitoring of FoG is essential for timely clinical intervention but remains challenging due to high power consumption, annotation cost, and the controlled environments required by current wearables. We introduce LIFT-PD (the source code is available at: https://github.com/shovito66/LIFT-PD ), a novel self-supervised learning (SSL) framework for real-time, patient-independent FoG detection that uniquely utilizes a single waist-worn accelerometer—an approach traditionally considered less optimal due to weaker gait signatures. LIFT-PD leverages SSL on unlabeled data collected from uncontrolled, real-world settings and employs a novel Differential Hopping Windowing Technique (DHWT) to address gait variability and dataset imbalance. Additionally, an opportunistic inference module selectively activates the deep learning model only during patient movement, significantly reducing power consumption and enabling continuous monitoring ( \(>\) 48 hours). Experimental results show that LIFT-PD achieves a 7.25% increase in precision and 4.4% improvement in accuracy compared to supervised and semi-supervised baseline models while requiring approximately 40% fewer labeled training samples. Evaluations across diverse patient characteristics-including severity, medication state, age, and gender-confirm the model’s robustness and clinical applicability, positioning LIFT-PD as a practical, energy-efficient, and scalable solution for continuous real-world FoG monitoring in PD.- 16 May 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Computing for Healthcare
- Vol. 7 (3) , 1-24
Open AccessSelf-Supervised Learning and Opportunistic Inference for Continuous Monitoring of Freezing of Gait in Parkinson’s Disease
Parkinson’s disease (PD) significantly affects patients’ quality of life through debilitating motor symptoms, such as Freezing of Gait (FoG). Continuous, in-home monitoring of FoG is essential for timely clinical intervention but remains challenging due to high power consumption, annotation cost, and the controlled environments required by current wearables. We introduce LIFT-PD (the source code is available at: https://github.com/shovito66/LIFT-PD ), a novel self-supervised learning (SSL) framework for real-time, patient-independent FoG detection that uniquely utilizes a single waist-worn accelerometer—an approach traditionally considered less optimal due to weaker gait signatures. LIFT-PD leverages SSL on unlabeled data collected from uncontrolled, real-world settings and employs a novel Differential Hopping Windowing Technique (DHWT) to address gait variability and dataset imbalance. Additionally, an opportunistic inference module selectively activates the deep learning model only during patient movement, significantly reducing power consumption and enabling continuous monitoring ( \(>\) 48 hours). Experimental results show that LIFT-PD achieves a 7.25% increase in precision and 4.4% improvement in accuracy compared to supervised and semi-supervised baseline models while requiring approximately 40% fewer labeled training samples. Evaluations across diverse patient characteristics-including severity, medication state, age, and gender-confirm the model’s robustness and clinical applicability, positioning LIFT-PD as a practical, energy-efficient, and scalable solution for continuous real-world FoG monitoring in PD. - Top Cited PapersOpen Access
- 1 March 2026
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 27 (2)
Large language model agents for biological intelligence across genomics, proteomics, spatial biology, and biomedicine
Large language models (LLMs) are evolving from passive predictors into agentic systems capable of planning, tool-use, and multimodal reasoning. This shift is especially consequential for biology, where complex, noisy, and multi-scale data require adaptive and integrative computational strategies. In this review, we provide the first systematic synthesis of LLM-based agents across genomics, molecular biology, imaging, biomedical analysis, and automated bioinformatics workflows. We analyze >60 emerging systems and organize them within a unifying framework that characterizes agentic traits, such as autonomous decision-making, external tool invocation, memory, and self-correction. Across domains, agentic LLMs show early promise in enabling multi-step analysis, linking heterogeneous evidence, and supporting exploratory scientific tasks. At the same time, our comparative assessment highlights consistent challenges, including unstable reasoning, limited biological grounding, retrieval misalignment, and barriers to reproducibility and biosafety. We conclude by outlining opportunities for trustworthy and collaborative biological agents, including multimodal integration, closed-loop experimental design, and robust evaluation practices. This survey aims to clarify the emerging landscape and chart a path toward reliable agentic systems for biological discovery.
- 1 March 2026
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 27 (2)
Top Cited PapersOpen AccessLarge language model agents for biological intelligence across genomics, proteomics, spatial biology, and biomedicine
Large language models (LLMs) are evolving from passive predictors into agentic systems capable of planning, tool-use, and multimodal reasoning. This shift is especially consequential for biology, where complex, noisy, and multi-scale data require adaptive and integrative computational strategies. In this review, we provide the first systematic synthesis of LLM-based agents across genomics, molecular biology, imaging, biomedical analysis, and automated bioinformatics workflows. We analyze >60 emerging systems and organize them within a unifying framework that characterizes agentic traits, such as autonomous decision-making, external tool invocation, memory, and self-correction. Across domains, agentic LLMs show early promise in enabling multi-step analysis, linking heterogeneous evidence, and supporting exploratory scientific tasks. At the same time, our comparative assessment highlights consistent challenges, including unstable reasoning, limited biological grounding, retrieval misalignment, and barriers to reproducibility and biosafety. We conclude by outlining opportunities for trustworthy and collaborative biological agents, including multimodal integration, closed-loop experimental design, and robust evaluation practices. This survey aims to clarify the emerging landscape and chart a path toward reliable agentic systems for biological discovery.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Engineering in Medicine and Biology
- Vol. 7, 241-249
Time-Aware Multimodal Sensor Fusion for Blood Glucose Forecasting in Non-Diabetic Adults
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Engineering in Medicine and Biology
- Vol. 7, 241-249
Open AccessTime-Aware Multimodal Sensor Fusion for Blood Glucose Forecasting in Non-Diabetic Adults
- 15 October 2025
- Published by Elsevier in Current Opinion in Biomedical Engineering
- Vol. 36
AI-powered wearable sensors for health monitoring and clinical decision making
- 15 October 2025
- Published by Elsevier in Current Opinion in Biomedical Engineering
- Vol. 36
AI-powered wearable sensors for health monitoring and clinical decision making
- Open Access
- 10 October 2025
- Published by MDPI AG in Preprints.org
AI-Powered Wearable Sensors for Health Monitoring and Clinical Decision Making
AI-powered wearable sensors are transforming remote health monitoring by enabling real-time diagnostics, personalized interventions and proactive disease management. This review synthesizes recent advances in AI-integrated biosensors across conditions such as diabetes, cardiovascular disease, neurodegenerative dis- orders, mental health, and maternal/neonatal care, while addressing challenges of scalability, privacy, interoperability, and model robustness. We highlight machine learning methods—including federated learning, transfer learning, and edge- AI—that enhance the processing of physiological signals i.e., glucose levels, gait patterns, and heart rate variability. Key innovations, including FDA-approved glucose monitors and digital twins for predictive health modeling, underscore the shift toward patient-centric and data-driven care. Yet, persistent gaps remain, including device heterogeneity, privacy concerns, and the need for adaptive models that generalize across populations. Emerging approaches such as large language models and counterfactual explanations provide contextualized insights and transparent decision-making. By bridging technical advances with clinical needs, this review charts a roadmap toward democratized, equitable, and precise healthcare.
- 10 October 2025
- Published by MDPI AG in Preprints.org
Open AccessAI-Powered Wearable Sensors for Health Monitoring and Clinical Decision Making
AI-powered wearable sensors are transforming remote health monitoring by enabling real-time diagnostics, personalized interventions and proactive disease management. This review synthesizes recent advances in AI-integrated biosensors across conditions such as diabetes, cardiovascular disease, neurodegenerative dis- orders, mental health, and maternal/neonatal care, while addressing challenges of scalability, privacy, interoperability, and model robustness. We highlight machine learning methods—including federated learning, transfer learning, and edge- AI—that enhance the processing of physiological signals i.e., glucose levels, gait patterns, and heart rate variability. Key innovations, including FDA-approved glucose monitors and digital twins for predictive health modeling, underscore the shift toward patient-centric and data-driven care. Yet, persistent gaps remain, including device heterogeneity, privacy concerns, and the need for adaptive models that generalize across populations. Emerging approaches such as large language models and counterfactual explanations provide contextualized insights and transparent decision-making. By bridging technical advances with clinical needs, this review charts a roadmap toward democratized, equitable, and precise healthcare.
- Open Access
- 29 July 2025
- Published by MDPI AG in Preprints.org
Trustworthy AI in Digital Health: A Comprehensive Review of Robustness and Explainability
Ensuring trust in AI systems is essential for the safe and ethical integration of machine learning systems into high-stakes domains such as digital health. Key dimensions, including robustness, explainability, fairness, accountability, and privacy, need to be addressed throughout the AI lifecycle, from problem formulation and data collection to model deployment and human interaction. While various contributions address different aspects of trustworthy AI, a focused synthesis on robustness and explainability, especially tailored to the healthcare context, remains limited. This review addresses that need by organizing recent advancements into an accessible framework, highlighting both technical and practical considerations. We present a structured overview of methods, challenges, and solutions, aiming to support researchers and practitioners in developing reliable and explainable AI solutions for digital health. This review article is organized into three main parts. First, we introduce the pillars of trustworthy AI and discuss the technical and ethical challenges, particularly in the context of digital health. Second, we explore application-specific trust considerations across domains such as intensive care, neonatal health, and metabolic health, highlighting how robustness and explainability support trust. Lastly, we present recent advancements in techniques aimed at improving robustness under data scarcity and distributional shifts, as well as explainable AI methods ranging from feature attribution to gradient-based interpretations and counterfactual explanations. This paper is further enriched with detailed discussions of the contributions toward robustness and explainability in digital health, the development of trustworthy AI systems in the era of LLMs, and various evaluation metrics for measuring trust and related parameters such as validity, fidelity, and diversity.
- 29 July 2025
- Published by MDPI AG in Preprints.org
Open AccessTrustworthy AI in Digital Health: A Comprehensive Review of Robustness and Explainability
Ensuring trust in AI systems is essential for the safe and ethical integration of machine learning systems into high-stakes domains such as digital health. Key dimensions, including robustness, explainability, fairness, accountability, and privacy, need to be addressed throughout the AI lifecycle, from problem formulation and data collection to model deployment and human interaction. While various contributions address different aspects of trustworthy AI, a focused synthesis on robustness and explainability, especially tailored to the healthcare context, remains limited. This review addresses that need by organizing recent advancements into an accessible framework, highlighting both technical and practical considerations. We present a structured overview of methods, challenges, and solutions, aiming to support researchers and practitioners in developing reliable and explainable AI solutions for digital health. This review article is organized into three main parts. First, we introduce the pillars of trustworthy AI and discuss the technical and ethical challenges, particularly in the context of digital health. Second, we explore application-specific trust considerations across domains such as intensive care, neonatal health, and metabolic health, highlighting how robustness and explainability support trust. Lastly, we present recent advancements in techniques aimed at improving robustness under data scarcity and distributional shifts, as well as explainable AI methods ranging from feature attribution to gradient-based interpretations and counterfactual explanations. This paper is further enriched with detailed discussions of the contributions toward robustness and explainability in digital health, the development of trustworthy AI systems in the era of LLMs, and various evaluation metrics for measuring trust and related parameters such as validity, fidelity, and diversity.
- Open Access
Detection and Severity Assessment of Parkinson’s Disease Through Analyzing Wearable Sensor Data Using Gramian Angular Fields and Deep Convolutional Neural Networks
Parkinson’s disease (PD) is the second-most common neurodegenerative disease. With more than 20,000 new diagnosed cases each year, PD affects millions of individuals worldwide and is most prevalent in the elderly population. The current clinical methods for the diagnosis and severity assessment of PD rely on the visual and physical examination of subjects and identifying key disease motor signs and symptoms such as bradykinesia, rigidity, tremor, and postural instability. In the present study, we developed a method for the diagnosis and severity assessment of PD using Gramian Angular Fields (GAFs) in combination with deep Convolutional Neural Networks (CNNs). Our model was applied to PD gait signals captured using pressure sensors embedded into insoles. Our results indicated an accuracy of 98.6%, a true positive rate (TPR) of 99.2%, and a true negative rate (TNR) of 98.5%, showcasing superior classification performance for PD diagnosis compared to the methods used in recent studies in the literature. The estimation of disease severity scores using gait signals showed a high accuracy for the Hoehn and Yahr score as well as the Timed Up and Go (TUG) test score (R2 > 0.8), while we achieved a lower prediction performance for the Unified Parkinson’s Disease Rating Scale (UPDRS) and its motor component (UPDRSM) scores (R2 < 0.2). These results were achieved using gait signals recorded in time windows as small as 10 s, which may pave the way for shorter, more accessible assessment tools for diagnosis and severity assessment of PD.
Open AccessDetection and Severity Assessment of Parkinson’s Disease Through Analyzing Wearable Sensor Data Using Gramian Angular Fields and Deep Convolutional Neural Networks
Parkinson’s disease (PD) is the second-most common neurodegenerative disease. With more than 20,000 new diagnosed cases each year, PD affects millions of individuals worldwide and is most prevalent in the elderly population. The current clinical methods for the diagnosis and severity assessment of PD rely on the visual and physical examination of subjects and identifying key disease motor signs and symptoms such as bradykinesia, rigidity, tremor, and postural instability. In the present study, we developed a method for the diagnosis and severity assessment of PD using Gramian Angular Fields (GAFs) in combination with deep Convolutional Neural Networks (CNNs). Our model was applied to PD gait signals captured using pressure sensors embedded into insoles. Our results indicated an accuracy of 98.6%, a true positive rate (TPR) of 99.2%, and a true negative rate (TNR) of 98.5%, showcasing superior classification performance for PD diagnosis compared to the methods used in recent studies in the literature. The estimation of disease severity scores using gait signals showed a high accuracy for the Hoehn and Yahr score as well as the Timed Up and Go (TUG) test score (R2 > 0.8), while we achieved a lower prediction performance for the Unified Parkinson’s Disease Rating Scale (UPDRS) and its motor component (UPDRSM) scores (R2 < 0.2). These results were achieved using gait signals recorded in time windows as small as 10 s, which may pave the way for shorter, more accessible assessment tools for diagnosis and severity assessment of PD.
- Open Access
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (28) , 29495-29497
Domain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract)
FuSE-MET addresses critical challenges in deploying human activity recognition (HAR) systems in uncontrolled environments by effectively managing noisy labels, sparse data, and undefined activity vocabularies. By integrating BERT-based word embeddings with domain-specific knowledge (i.e., MET values), FuSE-MET optimizes label merging, reducing label complexity and improving classification accuracy. Our approach outperforms the state-of-the-art techniques, including ChatGPT-4, by balancing semantic meaning and physical intensity.
- 11 April 2025
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 39 (28) , 29495-29497
Open AccessDomain-Informed Label Fusion Surpasses LLMs in Free-Living Activity Classification (Student Abstract)
FuSE-MET addresses critical challenges in deploying human activity recognition (HAR) systems in uncontrolled environments by effectively managing noisy labels, sparse data, and undefined activity vocabularies. By integrating BERT-based word embeddings with domain-specific knowledge (i.e., MET values), FuSE-MET optimizes label merging, reducing label complexity and improving classification accuracy. Our approach outperforms the state-of-the-art techniques, including ChatGPT-4, by balancing semantic meaning and physical intensity.