Bio

Dr. Schmiedmayer is an Instructor at the Stanford School of Medicine’s Division of Computational Medicine. His research sits at the intersection of artificial intelligence, computational medicine, and software engineering, where he develops scalable, open-source agentic AI systems and multimodal foundation models that integrate electronic health records, wearable sensors, connected devices, and patient-generated data. His work focuses on translating these technologies into patient-centered, privacy-preserving healthcare solutions, with applications in cardiometabolic disease prevention, personalized medicine, and clinical decision support.

His research addresses fundamental challenges in computational medicine by developing interoperable AI infrastructure and methodologies that enable the secure, scalable deployment of intelligent healthcare systems in both research and clinical practice. Through open-source software, he aims to accelerate the translation of AI innovations into real-world patient care.

He earned his doctoral degree (Dr. rer. nat.; the German equivalent of a PhD in computer science) from the Technical University of Munich, where he studied software engineering, mobile systems including smart devices, the application and integration of machine learning techniques, and the evolution of web service–based distributed systems. He holds a master’s and bachelor’s degree in computer science from the Technical University of Munich.

Academic Appointments

Professional Education

  • Dr. rer. nat., Technical University of Munich, (German equivalent of a PhD in computer science) (2022)
  • M.Sc., Technical University of Munich, Informatics (2019)
  • B.Sc., Technical University of Munich, Informatics (2017)

Additional Info

Current Research and Scholarly Interests

Artificial intelligence is fundamentally reshaping healthcare, yet most AI systems remain isolated prototypes that fail to integrate seamlessly into clinical workflows or empower patients in their everyday lives. Advancing computational medicine requires more than increasingly capable AI models; it demands interoperable, open, and patient-centered systems that securely integrate multimodal health data and translate AI innovations into routine clinical practice.

Dr. Schmiedmayer’s research develops the next generation of open-source agentic AI systems and multimodal foundation models for computational medicine. His work integrates electronic health records, wearable sensors, connected devices, medical knowledge, and patient-generated data to create intelligent systems that support personalized healthcare, clinical decision-making, and early disease prevention. A particular focus of his research is cardiometabolic disease prevention, where AI can help identify disease risk earlier, improve patient understanding, and enable personalized interventions before disease progression.

A central research direction is developing multimodal foundation models that can reason over longitudinal patient data. By combining structured health records with continuous time-series data from wearables, smartphones, and connected medical devices, his research investigates time-series language models that bridge heterogeneous data modalities and enable agentic AI to understand patient trajectories, identify digital biomarkers, and provide context-aware, personalized support. These advances enable patient-facing AI systems that continuously adapt to an individual’s health status while preserving privacy and operating reliably in real-world settings.

By combining advances in software engineering, artificial intelligence, multimodal machine learning, and computational medicine, Dr. Schmiedmayer’s long-term goal is to establish the open infrastructure that enables trustworthy, patient-centered AI to become a routine component of healthcare. His research aims to accelerate the translation of AI from research laboratories into scalable clinical systems that improve disease prevention, personalize care, and enable proactive healthcare.

All Publications

  • Spezi Data Pipeline: Streamlining FHIR-based interoperable digital dealth data workflows. NPJ digital medicine Bikia, V., Schmiedmayer, P., Zahedivash, A., Aalami, L., Rao, A., Ravi, V., Turk, M., Ceresnak, S. R., Aalami, O. 2026

    Abstract

    The increasing adoption of digital health technologies has amplified the need for robust, interoperable solutions to manage complex healthcare data. We present the Spezi Data Pipeline, an open-source Python toolkit designed to streamline the analysis of digital health data, from secure access and retrieval through processing, visualization, and export. The Pipeline is integrated into the larger Stanford Spezi open-source ecosystem for developing research and translational digital health software systems. Leveraging Health Level Seven (HL7) Fast Healthcare Interoperability Resources (FHIR)-based data representations, the Pipeline enables standardized handling of diverse data types, including sensor-derived observations, electrocardiogram (ECG) recordings, and clinical questionnaires-across research and clinical environments. We detail the modular system architecture and demonstrate its application using real-world data from the Pediatric Apple Watch Study (PAWS) at Stanford University, in which the Pipeline facilitated efficient extraction, transformation, and clinician-driven review of Apple Watch ECG data, supporting annotation and comparative analysis alongside traditional monitors. By reducing the need for bespoke data engineering and enabling prospective, clinician-in-the-loop analysis within standardized workflows, the Spezi Data Pipeline supports reproducible and interoperable clinical research using routinely collected digital health data.

    View details for DOI 10.1038/s41746-026-02666-7

    View details for PubMedID 42120671

  • Fine-tuning LLMs in behavioral psychology for scalable health coaching. NPJ cardiovascular health Mantena, S., Johnson, A., Oppezzo, M., Schütz, N., Tolas, A., Doijad, R., Mattson, C. M., Lawrie, A., Ramirez-Posada, M., Schmiedmayer, P., Linos, E., King, A. C., Rodriguez, F., Kim, D. S., Ashley, E. A. 2025; 2 (1): 48

    Abstract

    Personalized, smartphone-based coaching improves physical activity but relies on static, human-crafted messages. We introduce My Heart Counts (MHC)-Coach, a large language model fine-tuned on the Transtheoretical Model of Change. MHC-Coach generates messages tailored to an individual's psychology (their "stage of change"), providing personalized support to foster long-term physical activity behavior change. To evaluate MHC-Coach's efficacy, 632 participants compared human-expert and MHC-Coach interventions encouraging physical activity. Among messages matched to an individual's stage of change, 68.0% (N = 430) preferred MHC-Coach-generated messages (P < 0.001). Blinded behavioral science experts (N = 2) rated MHC-Coach messages higher than human-expert messages for perceived effectiveness (4.4 vs. 2.8) and Transtheoretical Model alignment (4.1 vs. 3.5) on a 5-point Likert scale. This work demonstrates how language models can operationalize behavioral science frameworks for personalized health coaching, showing the potential for promoting long-term physical activity and reducing cardiovascular disease risk at scale.

    View details for DOI 10.1038/s44325-025-00083-5

    View details for PubMedID 40994879

    View details for PubMedCentralID PMC12454129

  • Dynamic fog computing for enhanced LLM execution in medical applications SMART HEALTH Zagar, P., Ravi, V., Aalami, L., Krusche, S., Aalami, O., Schmiedmayer, P. 2025; 36

    View details for DOI 10.1016/j.smhl.2025.100577

    View details for Web of Science ID 001612129900016

  • LLMonFHIR: A Physician-Validated, Large Language Model-Based Mobile Application for Querying Patient Electronic Health Data. JACC. Advances Schmiedmayer, P., Rao, A., Zagar, P., Aalami, L., Ravi, V., Zahedivash, A., Yao, D. H., Fereydooni, A., Aalami, O. 2025; 4 (6 Pt 1): 101780

    Abstract

    To improve healthcare quality and empower patients, federal legislation requires nationwide interoperability of electronic health records (EHRs) through Fast Healthcare Interoperability Resources (FHIR) application programming interfaces. Nevertheless, key barriers to patient EHR access-limited functionality, English, and health literacy-persist, impeding equitable access to these benefits.This study aimed to develop and evaluate a digital health solution to address barriers preventing patient engagement with personal health information, focusing on individuals managing chronic cardiovascular conditions.We present LLMonFHIR, an open-source mobile application that uses large language models (LLMs) to allow users to "interact" with their health records at any degree of complexity, in various languages, and with bidirectional text-to-speech functionality. In a pilot evaluation, physicians assessed LLMonFHIR responses to queries on 6 SyntheticMass FHIR patient datasets, rating accuracy, understandability, and relevance on a 5-point Likert scale.A total of 210 LLMonFHIR responses were evaluated by physicians, receiving high median scores for accuracy (5/5), understandability (5/5), and relevance (5/5). Challenges summarizing health conditions and retrieving lab results were noted, with variability in responses and occasional omissions underscoring the need for precise preprocessing of data.LLMonFHIR's ability to generate responses in multiple languages and at varying levels of complexity, along with its bidirectional text-to-speech functionality, give it the potential to empower individuals with limited functionality, English, and health literacy to access the benefits of patient-accessible EHRs.

    View details for DOI 10.1016/j.jacadv.2025.101780

    View details for PubMedID 40373519

  • Enhancing Distress Tolerance Skills in Adolescents With Anorexia Nervosa Through the BALANCE Mobile App: Feasibility and Acceptability Study. JMIR formative research Miranda, C., Matheson, B., Datta, N., Whyte, A., Yang, H. J., Schmiedmayer, P., Ravi, V., Aalami, O., Lock, J. 2025; 9: e70278

    Abstract

    Anorexia nervosa is a severe psychiatric disorder with high morbidity and mortality, particularly among adolescents. Family-based treatment (FBT) is the leading evidence-based intervention for adolescent anorexia nervosa, involving parents in renourishment and behavior interruption. Despite its effectiveness, challenges in distress tolerance and emotion regulation during high-stress situations, such as mealtimes, contribute to suboptimal treatment outcomes, with only 35% to 50% of adolescents achieving full recovery. Enhancing distress tolerance skills during FBT may improve treatment responses and recovery rates. The BALANCE mobile app was developed to address this need, offering real-time, dialectical behavior therapy (DBT)-based distress tolerance skills to support adolescents and families during mealtimes.Our aim was to explore the feasibility and acceptability of a mobile app designed to deliver distress tolerance skills to adolescents with and adolescents without anorexia nervosa. When fully programmed and optimized, we plan to use the mobile app to improve distress tolerance during mealtimes for adolescents with anorexia nervosa undergoing FBT.BALANCE was developed collaboratively with Stanford University's Center for Biodesign, leveraging the expertise of clinical psychologists and using biodesign student input and the Stanford Spezi ecosystem. The app underwent an iterative development process, with feedback from adolescent users. The initial feasibility and acceptability of the app were assessed through self-reported questionnaires and structured interviews with 24 adolescents aged 12 to 18 years, including 4 diagnosed with anorexia nervosa and 20 healthy controls. Adolescents with anorexia nervosa specifically used the app during mealtimes, and healthy controls used it as needed. Participants assessed the app's usability, perceived effectiveness, and its impact on their distress tolerance.The app demonstrated high usability and acceptability. Of 24 participants, 83% (n=20) reported enjoying the app, 88% (n=21) would recommend it to peers, and 100% (n=24) found it user-friendly. Adolescents with anorexia nervosa reported that BALANCE helped them manage stressful mealtimes more effectively, highlighting features such as guided meditation, breathing exercises, and gamification elements as particularly effective. Healthy controls provided additional feedback, confirming the app's broad appeal to the target audience and potential scalability. Preliminary findings suggest that BALANCE may enhance distress tolerance in adolescents with and adolescents without anorexia nervosa.BALANCE shows promise as an innovative mobile health intervention for enhancing distress tolerance in adolescents with anorexia nervosa. Its user-friendly design and tailored DBT-based skills make it a feasible tool for integration into FBT. Future research should explore its integration into clinical practice and its impact on treatment outcomes. As distress tolerance skills are relevant to a range of mental health conditions, future research may also expand BALANCE's application to broader adolescent populations.

    View details for DOI 10.2196/70278

    View details for PubMedID 40019817

  • GPTCoach: Towards LLM-Based Physical Activity Coaching Jorke, M., Sapkota, S., Warkenthien, L., Vainio, N., Schmiedmayer, P., Brunskill, E., Landay, J. A., ACM ASSOC COMPUTING MACHINERY. 2025

    View details for DOI 10.1145/3706598.3713819

    View details for Web of Science ID 001501406100316

  • Spatial Continuity: Investigating Use Cases of Spatial Computing for Users with Low Vision Heidekrueger, P., Bruegge, B., Aalami, O., Schmiedmayer, P., ACM ASSOC COMPUTING MACHINERY. 2025

    View details for DOI 10.1145/3663547.3759711

    View details for Web of Science ID 001609881500158

  • Comprehensive real time remote monitoring for Parkinson's disease using Quantitative DigitoGraphy. NPJ Parkinson's disease Hoffman, S. L., Schmiedmayer, P., Gala, A. S., Wilkins, K. B., Parisi, L., Karjagi, S., Negi, A. S., Revlock, S., Coriz, C., Revlock, J., Ravi, V., Bronte-Stewart, H. 2024; 10 (1): 137

    Abstract

    People with Parkinson's disease (PWP) face critical challenges, including lack of access to neurological care, inadequate measurement and communication of motor symptoms, and suboptimal medication management and compliance. We have developed QDG-Care: a comprehensive connected care platform for Parkinson's disease (PD) that delivers validated, quantitative metrics of all motor signs in PD in real time, monitors the effects of adjusting therapy and medication adherence and is accessible in the electronic health record. In this article, we describe the design and engineering of all components of QDG-Care, including the development and utility of the QDG Mobility and Tremor Severity Scores. We present the preliminary results and insights from an at-home trial using QDG-Care. QDG technology has enormous potential to improve access to, equity of, and quality of care for PWP, and improve compliance with complex time-critical medication regimens. It will enable rapid "Go-NoGo" decisions for new therapeutics by providing high-resolution data that require fewer participants at lower cost and allow more diverse recruitment.

    View details for DOI 10.1038/s41531-024-00751-w

    View details for PubMedID 39068150

    View details for PubMedCentralID PMC11283542

  • Quantitative DigitoGraphy: a Comprehensive Real-Time Remote Monitoring System for Parkinson's Disease. Research square Hoffman, S. L., Schmiedmayer, P., Gala, A. S., Wilkins, K. B., Parisi, L., Karjagi, S., Negi, A. S., Revlock, S., Coriz, C., Revlock, J., Ravi, V., Bronte-Stewart, H. 2024

    Abstract

    People with Parkinson's disease (PWP) face critical challenges, including lack of access to neurological care, inadequate measurement and communication of motor symptoms, and suboptimal medication management and compliance. We have developed QDG-Care: a comprehensive connected care platform for Parkinson's disease (PD) that delivers validated, quantitative metrics of all motor signs in PD in real time, monitors the effects of adjusting therapy and medication adherence and is accessible in the electronic health record. In this article, we describe the design and engineering of all components of QDG-Care, including the development and utility of the QDG Mobility and Tremor Severity Scores. We present the preliminary results and insights from the first at-home trial using QDG-Care. QDG technology has enormous potential to improve access to, equity of, and quality of care for PWP, and improve compliance with complex time-critical medication regimens. It will enable rapid "Go-NoGo" decisions for new therapeutics by providing high-resolution data that require fewer participants at lower cost and allow more diverse recruitment.

    View details for DOI 10.21203/rs.3.rs-3783294/v1

    View details for PubMedID 38343821

    View details for PubMedCentralID PMC10854288

  • Utility of smart watches for identifying arrhythmias in children. Communications medicine Zahedivash, A., Chubb, H., Giacone, H., Boramanand, N. K., Dubin, A. M., Trela, A., Lencioni, E., Motonaga, K. S., Goodyer, W., Navarre, B., Ravi, V., Schmiedmayer, P., Bikia, V., Aalami, O., Ling, X. B., Perez, M., Ceresnak, S. R. 2023; 3 (1): 167

    Abstract

    Arrhythmia symptoms are frequent complaints in children and often require a pediatric cardiology evaluation. Data regarding the clinical utility of wearable technologies are limited in children. We hypothesize that an Apple Watch can capture arrhythmias in children.We present an analysis of patients ≤18 years-of-age who had signs of an arrhythmia documented by an Apple Watch. We include patients evaluated at our center over a 4-year-period and highlight those receiving a formal arrhythmia diagnosis. We evaluate the role of the Apple Watch in arrhythmia diagnosis, the results of other ambulatory cardiac monitoring studies, and findings of any EP studies.We identify 145 electronic-medical-record identifications of Apple Watch, and find arrhythmias confirmed in 41 patients (28%) [mean age 13.8 ± 3.2 years]. The arrythmias include: 36 SVT (88%), 3 VT (7%), 1 heart block (2.5%) and wide 1 complex tachycardia (2.5%). We show that invasive EP study confirmed diagnosis in 34 of the 36 patients (94%) with SVT (2 non-inducible). We find that the Apple Watch helped prompt a workup resulting in a new arrhythmia diagnosis for 29 patients (71%). We note traditional ambulatory cardiac monitors were worn by 35 patients (85%), which did not detect arrhythmias in 10 patients (29%). In 73 patients who used an Apple Watch for recreational or self-directed heart rate monitoring, 18 (25%) sought care due to device findings without any arrhythmias identified.We demonstrate that the Apple Watch can record arrhythmia events in children, including events not identified on traditionally used ambulatory monitors.

    View details for DOI 10.1038/s43856-023-00392-9

    View details for PubMedID 38092993

    View details for PubMedCentralID 4937287

  • CardinalKit: open-source standards-based, interoperable mobile development platform to help translate the promise of digital health. JAMIA open Aalami, O., Hittle, M., Ravi, V., Griffin, A., Schmiedmayer, P., Shenoy, V., Gutierrez, S., Venook, R. 2023; 6 (3): ooad044

    Abstract

    Smartphone devices capable of monitoring users' health, physiology, activity, and environment revolutionize care delivery, medical research, and remote patient monitoring. Such devices, laden with clinical-grade sensors and cloud connectivity, allow clinicians, researchers, and patients to monitor health longitudinally, passively, and persistently, shifting the paradigm of care and research from low-resolution, intermittent, and discrete to one of persistent, continuous, and high resolution. The collection, transmission, and storage of sensitive health data using mobile devices presents unique challenges that serve as significant barriers to entry for care providers and researchers alike. Compliance with standards like HIPAA and GDPR requires unique skills and practices. These requirements make off-the-shelf technologies insufficient for use in the digital health space. As a result, budget, timeline, talent, and resource constraints are the largest barriers to new digital technologies. The CardinalKit platform is an open-source project addressing these challenges by focusing on reducing these barriers and accelerating the innovation, adoption, and use of digital health technologies. CardinalKit provides a mobile template application and web dashboard to enable an interoperable foundation for developing digital health applications. We demonstrate the applicability of CardinalKit to a wide variety of digital health applications across 18 innovative digital health prototypes.

    View details for DOI 10.1093/jamiaopen/ooad044

    View details for PubMedID 37485467

    View details for PubMedCentralID PMC10356573

  • Reducing the Impact of Breaking Changes to Web Service Clients During Web API Evolution Schmiedmayer, P., Bauer, A., Bruegge, B., IEEE IEEE COMPUTER SOC. 2023: 1-11

    View details for DOI 10.1109/MOBILSoft59058.2023.00008

    View details for Web of Science ID 001031673900001

  • Global Software Engineering in a Global Classroom Schmiedmayer, P., Chatley, R., Bernius, J., Krusche, S., Chaika, K., Krinkin, K., Bruegge, B., IEEE Comp Soc IEEE COMPUTER SOC. 2022: 113-121

    View details for DOI 10.1109/ICSE-SEET55299.2022.9794211

    View details for Web of Science ID 000850196300012

  • Transitioning to a Large-Scale Distributed Programming Course Schmiedmayer, P., Reimer, L., Jovanovic, M., Henze, D., Jonas, S. edited by Daun, M., Hochmuller, E., Krusche, S., Brugge, B., Tenbergen, B. IEEE. 2020: 256-261

    View details for DOI 10.1109/cseet49119.2020.9206239

    View details for Web of Science ID 000651583800033

https://orcid.org/0000-0002-8607-9148