David Patric Werner Rastall
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
David Patric Werner Rastall has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Claire Allen
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
2 shared publications - David E. Newman-Toker
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
2 shared publications - Leonardo Eleuterio Ariello
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
2 shared publications - Mengchen Suo
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD
2 shared publications - Amir Kheradmand
Department of Neurology Johns Hopkins University School of Medicine Baltimore Maryland USA
1 shared publication - Daniel R. Gold
The Johns Hopkins University School of Medicine
1 shared publication
Latest Publications
- 8 September 2026
- Published by Wolters Kluwer Health in Neurology
- Vol. 107 (5) , e218381
Teaching Video NeuroImage: Lid Closure Unmasks Oculopalatal Tremor
- 8 September 2026
- Published by Wolters Kluwer Health in Neurology
- Vol. 107 (5) , e218381
Teaching Video NeuroImage: Lid Closure Unmasks Oculopalatal Tremor
- 7 September 2026
- Published by Elsevier in Neurologic Clinics
- Vol. 44 (4) , 685-701
Artificial Intelligence and Machine Learning for Eye Movements in Dizziness and Vertigo
- 7 September 2026
- Published by Elsevier in Neurologic Clinics
- Vol. 44 (4) , 685-701
Artificial Intelligence and Machine Learning for Eye Movements in Dizziness and Vertigo
- 1 September 2026
- Published by Wolters Kluwer Health in Neurology Open Access
- Vol. 2 (3)
Prevalence of Vestibular Dysfunction in Adult-Onset Cerebellar Ataxia
- 1 September 2026
- Published by Wolters Kluwer Health in Neurology Open Access
- Vol. 2 (3)
Prevalence of Vestibular Dysfunction in Adult-Onset Cerebellar Ataxia
- Open Access
Physiology-Driven Inference Using Large Language Models Enables Probabilistic Assessment of Huntington’s Disease from Smartphone Eye-Movement Data
Background: Artificial intelligence in medicine has largely relied on supervised training of disease-specific models, limiting scalability in conditions where labeled data are scarce. Large language models (LLMs), which encode broad medical knowledge through large-scale pretraining, offer an alternative paradigm in which structured physiological measurements can be interpreted directly without task-specific model training. Objective: To evaluate whether smartphone-derived ocular motor biomarkers can be translated into clinically meaningful probabilistic assessments of Huntington’s disease (HD) using general-purpose LLMs operating as inference engines. Methods: In this prospective proof-of-concept study, 26 participants (13 with genetically confirmed HD and 13 age-matched controls) completed a standardized ocular motor assessment using a custom smartphone application. Quantitative eye-movement metrics were validated against expert neurologist ratings. Structured physiological features were then provided to four general-purpose LLMs without task-specific training or diagnostic labels, and the models generated an AI-Assigned HD Probability Score (HAIPS). Discriminative performance and associations with clinical severity measures were evaluated. Results: Smartphone-derived ocular motor metrics showed strong agreement with clinician assessments (Spearman ρ = 0.76–0.95; all p < 0.001), confirming preservation of clinically meaningful physiological signals. LLM-derived HAIPS distinguished HD from controls with high accuracy (AUC 0.879–0.944), with no significant differences across models. Discrimination was statistically equivalent to a supervised logistic regression model trained on the same features. HAIPS correlated strongly with established measures of disease severity, including cognitive (MoCA, ρ = −0.86), functional (TFC, ρ = −0.74), and motor impairment (UHDRS, ρ = 0.85) (all p ≤ 0.003). Conclusions: Structured ocular motor biomarkers acquired using a consumer smartphone can be translated into clinically meaningful probabilistic assessments of HD by general-purpose LLMs without disease-specific model training. These findings support a framework in which physiologically grounded digital biomarkers are coupled with general-purpose inference models, potentially enabling scalable assessment in rare neurological diseases where labeled data are limited.
Open AccessPhysiology-Driven Inference Using Large Language Models Enables Probabilistic Assessment of Huntington’s Disease from Smartphone Eye-Movement Data
Background: Artificial intelligence in medicine has largely relied on supervised training of disease-specific models, limiting scalability in conditions where labeled data are scarce. Large language models (LLMs), which encode broad medical knowledge through large-scale pretraining, offer an alternative paradigm in which structured physiological measurements can be interpreted directly without task-specific model training. Objective: To evaluate whether smartphone-derived ocular motor biomarkers can be translated into clinically meaningful probabilistic assessments of Huntington’s disease (HD) using general-purpose LLMs operating as inference engines. Methods: In this prospective proof-of-concept study, 26 participants (13 with genetically confirmed HD and 13 age-matched controls) completed a standardized ocular motor assessment using a custom smartphone application. Quantitative eye-movement metrics were validated against expert neurologist ratings. Structured physiological features were then provided to four general-purpose LLMs without task-specific training or diagnostic labels, and the models generated an AI-Assigned HD Probability Score (HAIPS). Discriminative performance and associations with clinical severity measures were evaluated. Results: Smartphone-derived ocular motor metrics showed strong agreement with clinician assessments (Spearman ρ = 0.76–0.95; all p < 0.001), confirming preservation of clinically meaningful physiological signals. LLM-derived HAIPS distinguished HD from controls with high accuracy (AUC 0.879–0.944), with no significant differences across models. Discrimination was statistically equivalent to a supervised logistic regression model trained on the same features. HAIPS correlated strongly with established measures of disease severity, including cognitive (MoCA, ρ = −0.86), functional (TFC, ρ = −0.74), and motor impairment (UHDRS, ρ = 0.85) (all p ≤ 0.003). Conclusions: Structured ocular motor biomarkers acquired using a consumer smartphone can be translated into clinically meaningful probabilistic assessments of HD by general-purpose LLMs without disease-specific model training. These findings support a framework in which physiologically grounded digital biomarkers are coupled with general-purpose inference models, potentially enabling scalable assessment in rare neurological diseases where labeled data are limited.
- 17 July 2025
- Published by Springer Nature in Journal of Medical Systems
- Vol. 49 (1) , 98
Why Clinical Trials Will Fail to Ensure Safe AI
Recent reports have raised concerns about emergent behaviors in next-generation artificial intelligence (AI) models. These systems have been documented selectively adapting their behaviors during testing to falsify experimental outcomes and bypass regulatory oversight. This phenomenon-alignment faking-represents a fundamental challenge to medical AI safety. Regulatory strategies have largely adapted established protocols like clinical trials and medical device approval frameworks, but for next-generation AI these approaches may fail. This paper introduces alignment faking to a medical audience and critically evaluates how current regulatory tools are inadequate for advanced AI systems. We propose continuous logging through "AI SOAP notes" as a first step toward transparent and accountable AI functionality in clinical settings.
- 17 July 2025
- Published by Springer Nature in Journal of Medical Systems
- Vol. 49 (1) , 98
Why Clinical Trials Will Fail to Ensure Safe AI
Recent reports have raised concerns about emergent behaviors in next-generation artificial intelligence (AI) models. These systems have been documented selectively adapting their behaviors during testing to falsify experimental outcomes and bypass regulatory oversight. This phenomenon-alignment faking-represents a fundamental challenge to medical AI safety. Regulatory strategies have largely adapted established protocols like clinical trials and medical device approval frameworks, but for next-generation AI these approaches may fail. This paper introduces alignment faking to a medical audience and critically evaluates how current regulatory tools are inadequate for advanced AI systems. We propose continuous logging through "AI SOAP notes" as a first step toward transparent and accountable AI functionality in clinical settings.
- Open Access
- 20 April 2025
- Published by BMJ in BMJ Open Quality
- Vol. 14 (2) , e003135
Roads diverge: mapping the journey towards diagnostic health equity
Background: Diagnostic disparities significantly impact health outcomes, with various factors influencing differential access and treatment across communities. This paper explores the complex landscape of diagnosis and of variable diagnostic journeys, focusing on the multifaceted nature of diagnostic inequities and proposing a socioecological model to understand and address these disparities. Methods: The authors employed an interdisciplinary approach, synthesising insights from their respective knowledge domains. Through iterative discussions, visual diagramming and narrative development, a new framework was constructed to address the complexity of diagnostic inequities. Results: The ‘diagnostic environment’ model was developed, conceptualising diagnosis beyond the biomedical process to include three levels of influence: atmospheric (broad, pervasive societal factors), surface (direct healthcare interactions) and subterranean (underlying, often invisible societal norms and biases). This model integrates process, social phenomena and multilevel influences to more comprehensively address diagnostic disparities. Two additional frameworks: diagnostic journeys within this environment and issue constellations guiding navigation and intervention were developed. Conclusions: Diagnostic processes are influenced by a range of factors from individual to systemic levels, often leading to significant disparities in health outcomes. The proposed model offers a framework for understanding these disparities and suggests targeted interventions to address them. By integrating diverse perspectives and focusing on multilevel influences, the framework aims to improve diagnostic equity. Implications: This study highlights the need for a more holistic approach to diagnosis, acknowledging the complex interplay of factors that contribute to diagnostic disparities. The socioecological model provides a basis for future research and policy interventions aimed at reducing these disparities and improving overall diagnostic accuracy and equity.
- 20 April 2025
- Published by BMJ in BMJ Open Quality
- Vol. 14 (2) , e003135
Open AccessRoads diverge: mapping the journey towards diagnostic health equity
Background: Diagnostic disparities significantly impact health outcomes, with various factors influencing differential access and treatment across communities. This paper explores the complex landscape of diagnosis and of variable diagnostic journeys, focusing on the multifaceted nature of diagnostic inequities and proposing a socioecological model to understand and address these disparities. Methods: The authors employed an interdisciplinary approach, synthesising insights from their respective knowledge domains. Through iterative discussions, visual diagramming and narrative development, a new framework was constructed to address the complexity of diagnostic inequities. Results: The ‘diagnostic environment’ model was developed, conceptualising diagnosis beyond the biomedical process to include three levels of influence: atmospheric (broad, pervasive societal factors), surface (direct healthcare interactions) and subterranean (underlying, often invisible societal norms and biases). This model integrates process, social phenomena and multilevel influences to more comprehensively address diagnostic disparities. Two additional frameworks: diagnostic journeys within this environment and issue constellations guiding navigation and intervention were developed. Conclusions: Diagnostic processes are influenced by a range of factors from individual to systemic levels, often leading to significant disparities in health outcomes. The proposed model offers a framework for understanding these disparities and suggests targeted interventions to address them. By integrating diverse perspectives and focusing on multilevel influences, the framework aims to improve diagnostic equity. Implications: This study highlights the need for a more holistic approach to diagnosis, acknowledging the complex interplay of factors that contribute to diagnostic disparities. The socioecological model provides a basis for future research and policy interventions aimed at reducing these disparities and improving overall diagnostic accuracy and equity.
- 4 January 2025
- Published by Springer Nature in The Cerebellum
- Vol. 24 (1) , 1-6
Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia as a Presentation of a Novel DNMT1 Mutation
A 50-year-old woman with a 20-year history of gait instability presented with new-onset vertigo and oscillopsia. Examination revealed bilateral vestibular loss, cerebellar ataxia, sensory neuropathy, a “yes-yes” head tremor, nystagmus and a family history of a similar syndrome. Genetic testing for cerebellar ataxia with neuropathy and bilateral vestibular areflexia syndrome (RFC1) was negative, but whole exome sequencing identified a novel mutation in the DNA methyltransferase 1 (DNMT1) gene, broadening the differential diagnosis for this phenotype. Management was focused on symptomatic treatment and genetic counseling. This case highlights the importance of considering DNMT1 mutations in patients with a similar clinical presentation.
- 4 January 2025
- Published by Springer Nature in The Cerebellum
- Vol. 24 (1) , 1-6
Cerebellar Ataxia, Neuropathy, and Vestibular Areflexia as a Presentation of a Novel DNMT1 Mutation
A 50-year-old woman with a 20-year history of gait instability presented with new-onset vertigo and oscillopsia. Examination revealed bilateral vestibular loss, cerebellar ataxia, sensory neuropathy, a “yes-yes” head tremor, nystagmus and a family history of a similar syndrome. Genetic testing for cerebellar ataxia with neuropathy and bilateral vestibular areflexia syndrome (RFC1) was negative, but whole exome sequencing identified a novel mutation in the DNA methyltransferase 1 (DNMT1) gene, broadening the differential diagnosis for this phenotype. Management was focused on symptomatic treatment and genetic counseling. This case highlights the importance of considering DNMT1 mutations in patients with a similar clinical presentation.
- 12 November 2024
- Published by Wolters Kluwer Health in Neurology
- Vol. 103 (9) , e209957
Teaching NeuroImage: Alice in Wonderland Syndrome: Anatomic Correlation
- 12 November 2024
- Published by Wolters Kluwer Health in Neurology
- Vol. 103 (9) , e209957
Teaching NeuroImage: Alice in Wonderland Syndrome: Anatomic Correlation
- 19 October 2022
- Published by Elsevier in Journal of the Neurological Sciences
- Vol. 443, 120454
Deep learning in acute vertigo diagnosis
- 19 October 2022
- Published by Elsevier in Journal of the Neurological Sciences
- Vol. 443, 120454
Deep learning in acute vertigo diagnosis