About

Hersh Sagreiya, MD, is an Assistant Professor of Radiology at the Hospital of the University of Pennsylvania. He is actively involved in medical imaging, with clinical expertise in abdominal imaging and ultrasound. Dr. Sagreiya is a member of the active medical staff at Penn Presbyterian Medical Center, Chester County Hospital, and Pennsylvania Hospital, and serves as an attending at the Hospital of the University of Pennsylvania in the Department of Radiology. His research focuses on informatics and machine learning applications in medical imaging. He has contributed to the operationalization of artificial intelligence in thyroid ultrasound imaging and has published work on hepatic and abdominal adiposity assessment using machine learning on CT scans, as well as automated segmentation of optic nerve structures. Dr. Sagreiya's work also includes exploring the role of artificial intelligence in radiology malpractice risk mitigation and differentiating alveolar-interstitial syndromes using lung ultrasound. His educational background includes an undergraduate degree from Harvard College, where he graduated magna cum laude in Biochemical Sciences, and an MD from Stanford Medical School. His professional activities encompass clinical practice, research, and contributions to advancing AI applications in radiology.

Research topics

• Medicine
• Computer science
• Artificial intelligence
• Radiology
• Machine learning

Selected publications

• Prospective clinical study for assessing real-time objective pain using a signal processing algorithm in conjunction with wearable forehead sensors

  Journal of Clinical Anesthesia · 2026-05-22

  articleOpen access1st authorCorresponding

  Pain imposes an enormous personal and economic cost and is typically measured by patient self-report using subjective ratings such as the Numerical Rating Scale (NRS). However, subjective scales present limitations, and there is a need for objective assessment of patient pain. This study assessed the Real-time Objective Pain Assessment (ROPA) algorithm that uses optical spectrometry in conjunction with signal processing to capture continuous real-time objective pain measurements in a noninvasive, portable, and inexpensive way. In a prospective cohort study enrolling 130 women in labor receiving neuraxial analgesia, we tested the ROPA algorithm using two wearable sensors, one of the many commercially available devices (ForeSight Elite) and a novel forehead sensor developed by CereVu Medical. Patient pain rating using the NRS served as ground truth. The ForeSight Elite and CereVu devices respectively demonstrated a receiver operating characteristic area-under-the-curve of 0.95 and 0.93 for no/mild vs. severe pain, 0.90 and 0.86 for no/mild/moderate vs. severe pain, and 0.88 and 0.87 for no/mild vs. moderate/severe pain. Overall, we successfully distinguished between pain levels and plan further multi-center clinical validation.

  PublisherDOI

• Prediction of Chronic Obstructive Pulmonary Disease Using Machine Learning, Clinical Summary Notes, and Vital Signs: A Single-Center Retrospective Cohort Study in the United States

  Advances in respiratory medicine · 2026-01-07

  articleOpen access

  Introduction: Chronic obstructive pulmonary disease (COPD) is a leading cause of morbidity and mortality. Early identification and timely intervention for COPD exacerbations can reduce hospitalizations and complications, as well as improve patient outcomes. Methods: To develop and evaluate predictive models for COPD exacerbations using machine learning (ML), we performed a retrospective study using intensive care unit patient records. Records including 31,667 clinical notes and 10,489 vital signs were used to train and validate two machine learning models to predict COPD exacerbations in patients with known or suspected COPD. Predictive performance was evaluated for support vector machine, quadratic discriminant analysis, and adaptive boosting algorithms using area under the receiver operating characteristic curve (AUC). Results: The clinical note-based support vector machine model achieved an AUC of 0.81 and accuracy of 84.0% in predicting COPD exacerbations. Data from patient monitors and hospital information systems provided sufficient information for accurate prediction, demonstrating the utility of combining physiological signals with clinical text data. Discussion: Clinically available patient data and vital signs can effectively predict COPD exacerbations, potentially enabling earlier interventions, improved outcomes, and reduced healthcare burden. These findings suggest that integrating unstructured clinical notes with structured vital signs using ML frameworks may improve early detection of exacerbation risk, thus enabling appropriate patient counseling, triage, and treatment based on COPD severity.

  PublisherOA PDFDOI

• Automated characterization of abdominal MRI exams using deep learning

  Scientific Reports · 2025-07-25 · 2 citations

  articleOpen accessSenior author

  Advances in magnetic resonance imaging (MRI) have revolutionized disease detection and treatment planning. However, the growing volume and complexity of MRI data-along with heterogeneity in imaging protocols, scanner technology, and labeling practices-creates a need for standardized tools to automatically identify and characterize key imaging attributes. Such tools are essential for large-scale, multi-institutional studies that rely on harmonized data to train robust machine learning models. In this study, we developed convolutional neural networks (CNNs) to automatically classify three core attributes of abdominal MRI: pulse sequence type, imaging orientation, and contrast enhancement status. Three distinct CNNs with similar backbone architectures were trained to classify single image slices into one of 12 pulse sequences, 4 orientations, or 2 contrast classes. The models achieved high classification accuracies of 99.51%, 99.87%, and 99.99% for pulse sequence, orientation, and contrast, respectively. We applied Grad-CAM to visualize image regions influencing pulse sequence predictions and highlight relevant anatomical features. To enhance performance, we implemented a majority voting approach to aggregate slice-level predictions, achieving 100% accuracy at the volume level for all tasks. External validation using the Duke Liver Dataset demonstrated strong generalizability; after adjusting for class label mismatch, volume-level accuracies exceeded 96.9% across all classification tasks.

  PublisherOA PDFDOI

• Automation of Ultrasonographic Optic Nerve Sheath Diameter Measurement: A Scoping Review

  Journal of Neuroimaging · 2025-01-01 · 7 citations

  reviewOpen access

  Intracranial pressure (ICP) monitoring is a cornerstone of neurocritical care in managing severe brain injury. However, current invasive ICP monitoring methods carry significant risks, including infection and intracranial hemorrhage, and are contraindicated in certain clinical situations. Additionally, these methods are not universally available. Optic nerve sheath diameter (ONSD) measurement presents a promising noninvasive alternative for ICP monitoring, though its clinical adoption has been limited due to its operator dependence and inconsistencies in imaging acquisition and measurement techniques. Automating both ONSD image acquisition and measurement could enhance accuracy and reliability, thereby improving its utility as a noninvasive ICP estimation tool. A range of image analysis and machine learning (ML) techniques have been applied to address these challenges. In this paper, we provide a narrative review of the current literature on ONSD automation, examining the strengths and limitations of classical image analysis and ML models in improving ONSD-based ICP assessment.

  PublisherOA PDFDOI

• Reply to “Minimizing Medical Malpractice Risk for Radiologists Using Artificial Intelligence”

  American Journal of Roentgenology · 2025-11-12

  article1st authorCorresponding
  PublisherDOI

• Development of Machine-Learning–Based Algorithms for Remote Respiratory Disease Monitoring and Assessment

  Military Medicine · 2025-06-30

  articleSenior author

  INTRODUCTION: A patient's breathing pattern reflects their work of breathing, and it can be monitored in patients to assess changes in their respiratory condition. Although current diagnostic and management practices primarily rely on specialist clinical testing, there is a need to monitor patients in field and combat hospitals or in transit from the point of injury. This work aims to develop machine learning-based algorithms to facilitate remote respiratory disease monitoring and assessment for patients undergoing continuous positive airway pressure therapy. MATERIALS AND METHODS: Data were collected from 30 healthy adults, encompassing respiratory pressure, flow, and dynamic thoraco-abdominal circumferential measurements under three breathing conditions: normal, panting, and deep breathing. These data were sourced from the publicly available PhysioNet, with associated ethical approval and informed consent. Various machine learning models, including the random forest classifier, logistic regression, and support vector machine, were trained to predict the correct breathing condition. Leave-one-out and k-fold cross-validation were performed. RESULTS: The random forest classifier demonstrated the highest accuracy, which was 88% after incorporating breathing rate as a feature. These findings support the potential of AI-driven respiratory monitoring systems to transition respiratory assessments from traditional clinical settings to remote monitoring, enhancing accessibility and patient autonomy. Future work involves validating these models in a large prospective study involving individuals who are actively deployed, while exploring additional machine learning techniques. CONCLUSIONS: Determining the correct breathing pattern relates to the patient's work of breathing, a marker of clinical deterioration of key importance for deployed soldiers. Early and accurate detection of the patient's respiratory condition can potentially improve outcomes in resource-constrained settings, particularly field hospitals and injured soldiers in transit from the point of injury. This work supports the potential of AI-driven respiratory monitoring to enhance remote respiratory monitoring, allowing for timely intervention and better long-term care.

  PublisherDOI

• Generalizable CT Vision-Language Modeling for Population Health and Disease Risk

  medRxiv · 2025-07-03 · 1 citations

  preprintOpen access

  Generalizable foundation models for computed tomographic (CT) medical imaging data are emerging AI tools anticipated to vastly improve clinical workflow efficiency. However, existing models are typically trained within narrow imaging contexts, including limited anatomical coverage, contrast settings, and clinical indications. These constraints reduce their ability to generalize across the broad spectrum of real-world presentations encountered in volumetric CT imaging data. We introduce Percival, a vision-language foundation model trained on over 400,000 CT volumes and paired radiology reports from more than 50,000 participants enrolled in the Penn Medicine BioBank. Percival employs a dual-encoder architecture with a transformer-based image encoder and a BERT-style language encoder, aligned via symmetric contrastive learning. Percival was validated on over 20,000 participants imaging data encompassing over 100,000 CT volumes. In image-text recall tasks, Percival outperforms models trained on limited anatomical windows. To assess Percival's clinical knowledge, we evaluated the biologic, phenotypic and prognostic relevance using laboratory-wide, phenome-wide association studies and survival analyses, uncovering a rich latent structure aligned with physiological measurements and disease phenotypes.

  PublisherOA PDFDOI

• Editorial for “Preoperative Assessment of Extraprostatic Extension in Prostate Cancer Using an Interpretable Tabular Prior‐Data Fitted Network‐Based Radiomics Model From <scp>MRI</scp> ”

  Journal of Magnetic Resonance Imaging · 2025-09-13

  editorial
  PublisherDOI

• Automated Integration of AI Results into Radiology Reports Using Common Data Elements

  Journal of Imaging Informatics in Medicine · 2025-01-27 · 13 citations

  articleOpen access

  Integration of artificial intelligence (AI) into radiology practice can create opportunities to improve diagnostic accuracy, workflow efficiency, and patient outcomes. Integration demands the ability to seamlessly incorporate AI-derived measurements into radiology reports. Common data elements (CDEs) define standardized, interoperable units of information. This article describes the application of CDEs as a standardized framework to embed AI-derived results into radiology reports. The authors defined a set of CDEs for measurements of the volume and attenuation of the liver and spleen. An AI system segmented the liver and spleen on non-contrast CT images of the abdomen and pelvis, and it recorded their measurements as CDEs using the Digital Imaging and Communications in Medicine Structured Reporting (DICOM-SR) framework to express the corresponding labels and values. The AI system successfully segmented the liver and spleen in non-contrast CT images and generated measurements of organ volume and attenuation. Automated systems extracted corresponding CDE labels and values from the AI-generated data, incorporated CDE values into the radiology report, and transmitted the generated image series to the Picture Archiving and Communication System (PACS) for storage and display. This study demonstrates the use of radiology CDEs in clinical practice to record and transfer AI-generated data. This approach can improve communication among radiologists and referring providers, harmonize data to enable large-scale research efforts, and enhance the performance of decision support systems. CDEs ensure consistency, interoperability, and clarity in reporting AI findings across diverse healthcare systems.

  PublisherOA PDFDOI

• Hepatic and abdominal adiposity in type 2 diabetes as assessed with machine learning on CT scans

  medRxiv · 2025-12-18

  articleOpen accessSenior authorCorresponding

  Abstract Aims The distribution of abdominal adipose depots and their mechanistic links to type 2 diabetes remain incompletely understood. This study elucidated the relationship between type 2 diabetes presence and quantitative abdominal imaging traits, including hepatic steatosis, liver and spleen size, and adipose distribution, using unenhanced computed tomography (CT) scans from a large-scale, racially diverse, disease-focused medical biobank. Materials and Methods Deep learning algorithms were applied to abdominal CT scans to automatically quantify image-derived phenotypes, including spleen-hepatic attenuation difference (SHAD) for hepatic steatosis, liver and spleen volumes (LV and SV, respectively), visceral and subcutaneous adipose tissue (VAT and SAT, respectively), and visceral-to-subcutaneous fat ratio (VSR). Results Diabetic individuals demonstrated a greater degree of hepatic steatosis and central adiposity than those without diabetes. Liver attenuation was lower (47.6 vs. 52.4 Hounsfield units (HU); lower values indicate greater steatosis), SHAD was higher (−5.41 vs. −8.41 HU; more positive values indicate greater steatosis), and steatosis prevalence increased (38.4% vs. 21.4%) (all p&lt;2.2×10⁻¹⁶). VSR was also elevated (0.64 vs. 0.54, p=5.86×10⁻¹³). These trends remained significant after stratification by sex. Multivariate analyses revealed independent associations of diabetes with SHAD (OR 1.04), LV (OR 1.59), SV (OR 3.95), VAT (OR 1.23), SAT (OR 1.05), and VSR (OR 2.27), after adjusting for age, sex, race, and BMI. Conclusions Hepatic steatosis, hepatomegaly, and visceral adiposity on CT imaging are predictive of type 2 diabetes presence. Notably, VSR showed a stronger association with diabetes than BMI, underscoring how body-fat distribution, rather than overall adiposity, more accurately reflects metabolic disease risk.

  PublisherOA PDFDOI

Frequent coauthors

• Amar K. Das

  Guardant (United States)

  26 shared

• Karthik Ponnusamy

  Hamad Medical Corporation

  25 shared

• Narmadan A. Kumarasamy

  Hackensack University Medical Center

  25 shared

• Doris Chen

  Stanford University

  25 shared

• Yiren Chen

  Argonne National Laboratory

  25 shared

• Daniel L. Rubin

  Stanford University

  14 shared

• Alireza Akhbardeh

  The University of Texas Health Science Center at Houston

  12 shared

• Russ B. Altman

  Stanford University

  10 shared

Labs

• Hersh Sagreiya LabPI