Jerry Bonnell

Dr. Jerry Bonnell

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 June 2021 in Journal of Open Humanities Data
Verified
Journal Article
Open access
PDF file
SASusanna Allés-Torrent
Susanna Allés-Torrent
GRGimena del Rio Riande
Gimena del Rio Riande
JBJerry Bonnell
Jerry Bonnell
DSDieyun Song
Dieyun Song
NHNidia Hernández
Nidia Hernández

22 February 2024 in Biomedicines
Verified
Journal Article
Open access
PDF file
JBJerry Bonnell
Jerry Bonnell
OAOscar Alcezar
Oscar Alcezar
BWBrandon Watts
Brandon Watts
Peter BuchwaldPeter Buchwald
Peter Buchwald
MAMidhat Abdulreda
Midhat Abdulreda
MOMitsu Ogihara
Mitsu Ogihara
Background: Type 1 diabetes (T1D) is a devastating autoimmune disease, and its rising prevalence in the United States and around the world presents a critical problem in public health. While some treatment options exist for patients already diagnosed, individuals considered at risk for developing T1D and who are still in the early stages of their disease pathogenesis without symptoms have no options for any preventive intervention. This is because of the uncertainty in determining their risk level and in predicting with high confidence who will progress, or not, to clinical diagnosis. Biomarkers that assess one’s risk with high certainty could address this problem and will inform decisions on early intervention, especially in children where the burden of justifying treatment is high. Single omics approaches (e.g., genomics, proteomics, metabolomics, etc.) have been applied to identify T1D biomarkers based on specific disturbances in association with the disease. However, reliable early biomarkers of T1D have remained elusive to date. To overcome this, we previously showed that parallel multi-omics provides a more comprehensive picture of the disease-associated disturbances and facilitates the identification of candidate T1D biomarkers. Methods: This paper evaluated the use of machine learning (ML) using data augmentation and supervised ML methods for the purpose of improving the identification of salient patterns in the data and the ultimate extraction of novel biomarker candidates in integrated parallel multi-omics datasets from a limited number of samples. We also examined different stages of data integration (early, intermediate, and late) to assess at which stage supervised parametric models can learn under conditions of high dimensionality and variation in feature counts across different omics. In the late integration scheme, we employed a multi-view ensemble comprising individual parametric models trained over single omics to address the computational challenges posed by the high dimensionality and variation in feature counts across the different yet integrated multi-omics datasets. Results: the multi-view ensemble improves the prediction of case vs. control and finds the most success in flagging a larger consistent set of associated features when compared with chance models, which may eventually be used downstream in identifying a novel composite biomarker signature of T1D risk. Conclusions: the current work demonstrates the utility of supervised ML in exploring integrated parallel multi-omics data in the ongoing quest for early T1D biomarkers, reinforcing the hope for identifying novel composite biomarker signatures of T1D risk via ML and ultimately informing early treatment decisions in the face of the escalating global incidence of this debilitating disease.

1 April 2026 in JCO Clinical Cancer Informatics
Verified
Journal Article
Closed access
ANAkina Natori
Akina Natori
JBJerry Bonnell
Jerry Bonnell
VSVasileios Stathias
Vasileios Stathias
SFSara E. Fleszar-Pavlovic
Sara E. Fleszar-Pavlovic
MOMitsunori Ogihara
Mitsunori Ogihara
FPFrank J. Penedo
Frank J. Penedo
PURPOSE: Effective risk stratification in cancer survivorship requires handling longitudinal data characterized by multimodal inputs, irregular follow-up, and recurrent clinical events. This study evaluated the incremental value of integrating patient-reported outcomes (PROs) with electronic health record (EHR) data and identified optimal windowing strategies for machine learning–based prediction of adverse survivorship outcomes. PATIENTS AND METHODS: This study used a cohort of 25,592 cancer survivors followed for 36 months. Data from four domains were integrated: baseline measures, treatments, PROs, and health care utilization (emergency room visits and hospitalizations). Two classification models, LASSO and CATBOOST, were applied across modality combinations and five temporal representations of patient history: static early-phase (0-6 months), cumulative history, sliding windows (4- and 12-month), and a most-recent baseline. Performance was evaluated for predicting monthly health care utilization and patient-reported symptom burden using average precision (AP). SHapley Additive exPlanations (SHAP) analysis identified key predictors and characterized their evolving influence. RESULTS: For health care utilization, CATBOOST models trained on the full multimodal data set with time-windowed predictors achieved strong discrimination (AP = 0.207), outperforming static baselines by 27%. SHAP analyses emphasized dynamic contributions from recent utilization and treatment toxicity. For symptom burden, PRO integration was crucial, nearly doubling clinical-only performance (AP = 0.132 v 0.071), with longer historical context improving characterization of progressive functional decline and symptom severity. Flagging the top 10% of patients by predicted risk captured 51.7% of health care utilizations and 46.7% of symptom burden events. CONCLUSION: Adverse survivorship risk is dynamic and outcome-specific: acute health care utilization is best predicted by recent clinical momentum, while longitudinal patient-reported trends drive symptom burden. Implementing decoupled, dynamic windows provides a flexible framework for risk stratification and risk prediction beyond standard clinical heuristics, facilitating proactive, precision-based survivorship care.

10 March 2020 in Digital Scholarship in the Humanities
Verified
Journal Article
Closed access
JBJerry Bonnell
Jerry Bonnell
MOMitsunori Ogihara
Mitsunori Ogihara
Morphological adornment of text in Text Encoding Initiative (TEI) XML can be useful for studies in textual analysis. MorphAdorner is a principal tool for providing such functionality in English texts. However, its practical use is limited when the input XML contains branching text, e.g. when appears, as it modifies the input document. In such cases, preprocessing is required to obtain the desired results. This article introduces a new tool Hoshi with the purpose of determining how this issue can be best handled with minimal input modification and preprocessing needed. It also investigates whether parsing software available online can be used to supply morphological information that can be encoded in an output format like MorphAdorner, and whether such a tool can be developed to adorn text in other languages. Challenges include those posed by the target language, the current software available for providing morphological analysis in it, and the schema needed for encoding the results. Moreover, technical hurdles presented by segmented and branching text can complicate the alignment process, especially when the intent is to guarantee input document integrity. Our approach for handling these is presented, and the article ends by outlining future applications of Hoshi that can help to enhance TEI scholarship that prioritizes the use of morphological word metadata.

7 January 2026 in Translational Behavioral Medicine
Verified
Journal Article
Closed access
FPFrank J Penedo
Frank J Penedo
SFSara E Fleszar-Pavlovic
Sara E Fleszar-Pavlovic
ANAkina Natori
Akina Natori
AWAndrew Wang
Andrew Wang
VSVasileios Stathias
Vasileios Stathias
MOMitsunori Ogihara
Mitsunori Ogihara
Among the 18 million cancer survivors in the United States, many report significant unmet physical, psychosocial, and practical needs that existing healthcare systems struggle to address at scale. This conceptual perspective examines how artificial intelligence (AI)-powered platforms can serve as adjuncts to care by addressing unmet needs, supporting symptom management, and delivering psychoeducation and psychosocial support. Drawing on evidence from oncology AI and cancer survivorship, we consider how AI may improve the quality, personalization, and accessibility of care, distinguishing applications supported by prospective evidence from those that remain in early stages. Preliminary evidence suggests AI has the potential to deliver personalized psychoeducation, symptom monitoring, and psychosocial support as adjuncts to clinical care. To guide responsible development and implementation of these tools, we propose the PRECISION-AI Adjunctive Supportive Care Development Framework, a systematic model for developing, validating, and implementing AI in supportive oncology care. As a primary case example, we describe My Wellness Support, an agentic AI-enabled, risk-stratified supportive care platform designed to augment standard survivorship care, including its conceptual framework and development, ethical considerations, data protections, transparency, and bias mitigation. A proof-of-concept study in breast cancer patients is currently underway. Outcome data are not yet available, and no effectiveness or safety claims are made for the platform. AI in cancer survivorship has potential to expand access to survivorship support and improve health-related quality of life, but realizing this potential requires rigorous prospective validation, equitable implementation across diverse populations, and sustained collaboration among healthcare providers, developers, survivors, ethicists, and policymakers.