Greenwood, Jason D.

Background & Aims: Multi-target stool DNA (mt-sDNA) is an established stool-based colorectal cancer screening test, yet long-term performance is not well-described. We evaluated follow-up colonoscopy rates & quality, neoplasia yield, and repeat-testing adherence over 10 years with an integrated screening program. Methods: We assembled a retrospective cohort of mt-sDNA tests (September 2014–July 2025) within a multi-site health system. A validated large-language-model pipeline extracted colonoscopy, pathology, and quality endpoints from free-text reports. Detection outcomes (adenoma detection rate, advanced neoplasia, modified advanced adenoma, sessile serrated lesion (SSL), and negative exams) were calculated among high-quality exams; adenocarcinoma, bowel preparation adequacy, and cecal intubation rate (CIR) used all screening exams. We assessed colonoscopy completion ≤1 year after positive mt-sDNA, follow-up colonoscopy findings, findings by time-to-colonoscopy, and adherence to repeat mt-sDNA after negative result. Results: Among 148,624 tests in 110,410 individuals, positivity was 13.6%. Of 19,506 positives, 65.1% completed colonoscopy within 12 months (median 82 days). Colonoscopy after a positive mt-sDNA had high yield: advanced adenoma diagnosis was 30.7%; adenocarcinoma was 1.3%. Of patients who had colonoscopy within 3 years of a negative mt-sDNA test, 45.2% had adenoma/serrated lesion, 9.3% had advanced adenoma and 0.5% had colon cancer. Adenoma detection rate after positive mt-sDNA (>50%) exceeded specialty society quality benchmarks and increased over time; Longer intervals (>6 months) to colonoscopy after a positive test were associated with higher advanced adenoma (trend P=0.001).After an initial negative mt-sDNA, only 35.3% were screened again by mt-sDNA testing within 3 years. Conclusions: Over a decade,mt-sDNA screening achieved high colonoscopy qualitybut revealed critical gaps in the care cascade. One-third of positive tests lacked timely follow-up and nearly two-thirds of negative tests were not repeated on schedule, suggesting opportunities for system-level interventions.

This chapter provides a high-level overview of clinical informatics (CI) and artificial intelligence (AI) in healthcare. It covers foundational concepts such as electronic health records (EHRs) and data standards, current applications including Clinical Decision Support Systems (CDSS) and ambient scribing/documentation, and near-term directions for both clinicians and patients. The chapter also introduces how foundational CI systems and healthcare data are leveraged to develop AI solutions, ranging from traditional machine learning to large language models, while addressing key issues such as bias, generalizability, and regulatory considerations. Finally, the chapter examines a potential future in which decentralized AI models augment clinicians in their daily work, generate deeper clinical insights, and transform patient care through advanced predictive models and digital twins.

INTRODUCTION: Colorectal cancer remains a leading cause of cancer associated death in the United States and colonoscopy the primary screening strategy for prevention. Rates of adenomatous and serrated neoplasia detection are inversely associated with postcolonoscopy colorectal cancer. This crucial quality metric depends on accurate ascertainment of colorectal neoplasia findings from both endoscopy and histopathology records. We aimed to assess the feasibility of a random forest machine learning model to rapidly and accurately categorize colorectal neoplasia from electronic health record data. METHODS: A retrospective cohort study compared neoplasia detection rates among individuals undergoing colonoscopy at a large academic institution to develop a rule-based algorithm to categorize colorectal neoplasia from endoscopy reports and pathology systematized nomenclature of medicine – clinical terms. This cohort provided a large training set to develop a natural language processing system using a random forest approach to automatically classify unstructured pathology findings into adenoma, serrated, or advanced neoplasms. This system was manually validated through an independent holdout set. RESULTS: The training set comprised 35,953 unstructured pathology reports with matched systematized nomenclature of medicine – clinical terms from 95,188 unstructured colonoscopy reports. The final model was assessed on an independent holdout set of 337 manually annotated procedures obtaining an area under the receiver operating characteristic curve of 0.997 (confidence interval [CI] 0.994–1), 0.99 (CI 0.98–1), and 0.99 (CI 0.98–0.99) for prediction of adenoma, serrated, and advanced lesions, respectively. DISCUSSION: The random forest-based hybrid natural language processing system for classification of colonoscopy results was both accurate and explainable. NLP combined with effective machine learning algorithms can provide a scalable strategy for colonoscopy quality monitoring.

Abstract Objectives Recent work has demonstrated that automated fluorescence flow cytometry (FLC) is a potential alternative for the detection and quantification of Plasmodium parasites. The objective of this study was to apply this novel FLC method to detect and quantify Babesia parasites in venous blood and compare results to light microscopy and polymerase chain reaction methods. Methods An automated hematology/malaria analyzer (XN-31; Sysmex) was used to detect and quantify B microti–infected red blood cells from residual venous blood samples (n = 250: Babesia positive, n = 170; Babesia negative, n = 80). As no instrument software currently exists for Babesia, qualitative and quantitative machine learning (ML) algorithms were developed to facilitate analysis. Results Performance of the ML models was verified against the XN-31 software using P falciparum–infected samples. When applied to Babesia-infected samples, the qualitative ML model demonstrated an area under the curve (AUC) of 0.956 (sensitivity, 95.9%; specificity, 83.3%) relative to polymerase chain reaction. For valid scattergrams, the qualitive model achieved an AUC of 1.0 (sensitivity and specificity, 100%), while the quantitative model demonstrated an AUC of 0.986 (sensitivity, 94.4%; specificity, 100%). Conclusions This investigation demonstrates that Babesia parasites can be detected and quantified directly from venous blood using FLC. Although promising, opportunities remain to improve the general applicability of the method.

Continuity of care has been an identifying characteristic of family medicine since its inception and is an essential ingredient for high-functioning health care teams. Many benefits, including the quadruple aim of enhancing patient experience, improving population health, reducing costs, and improving care team well-being, are ascribed to continuity of care. In 2023, the Accreditation Council for Graduate Medical Education (ACGME) added two new continuity requirements—annual patient-sided continuity and annual resident-sided continuity—in family medicine training programs. This article reviews continuity of care as it applies to family medicine training programs. We discuss the various types of continuity and issues surrounding the measurement of continuity. A generally agreed upon definition of patient-sided and resident-sided continuity is presented to allow programs to begin to collect the necessary data. Especially within resident training programs, intricacies associated with maintaining continuity of care, such as empanelment, resident turnover, and scheduling, are discussed. The importance of right-sizing resident panels is highlighted, and a mechanism for accomplishing this is presented. The recent ACGME requirements represent a cultural shift from measuring resident experience based on volume to measuring resident continuity. This cultural shift forces family medicine training programs to adapt their various systems, policies, and procedures to emphasize continuity. We hope this manuscript’s review of several facets of contuinuity, some unique to training programs, helps programs ensure compliance with the ACGME requirements.

A woman in her late fifties was admitted to the Family Medicine Inpatient Service directly from Rheumatology clinic for polyarticular pain and erythema with concern for infection. She was taking immunosuppressant medications for a history of multiple autoimmune diseases. Examination showed increasing erythema and tenderness on the upper and lower extremity joints. Histologic evaluation, surgical evaluation, and cultures were consistent with mycobacterium haemophilum infection. Mycobacterium haemophilum is an uncommon opportunistic infection that usually affects immunocompromised patients. The patient was treated with a multi-drug antibiotic regimen for several months due to drug resistance. Although this opportunistic infection is not common it should be considered in the differential of immunocompromised patients with skin and articular symptoms. Treatment outcomes are usually favorable if it caught earlier in the course.