Machine learning in heart failure diagnosis, prediction, and prognosis: review.

Globally, cardiovascular diseases take the lives of over 17 million people each year, mostly through myocardial infarction, or MI, and heart failure (HF). This comprehensive literature review examines various aspects related to the diagnosis, prediction, and prognosis of HF in the context of machine learning (ML). The review covers an array of topics, including the diagnosis of HF with preserved ejection fraction (HFpEF) and the identification of high-risk patients with HF with reduced ejection fraction (HFrEF). The prediction of mortality in different HF populations using different ML approaches is explored, encompassing patients in the ICU, and HFpEF patients using biomarkers and gene expression. The review also delves into the prediction of mortality and hospitalization rates in HF patients with mid-range ejection fraction (HFmrEF) using ML methods. The findings highlight the significance of a multidimensional approach that encompasses clinical evaluation, laboratory assessments, and comprehensive research to improve our understanding and management of HF. Promising predictive models incorporating biomarkers, gene expression, and consideration of epigenetics demonstrate potential in estimating mortality and identifying high-risk HFpEF patients. This literature review serves as a valuable resource for researchers, clinicians, and healthcare professionals seeking a comprehensive and updated understanding of the role of ML diagnosis, prediction, and prognosis of HF across different subtypes and patient populations.

Assessment of Agenda Setting With ACGME Milestones in Family Medicine Residents.

BACKGROUND AND OBJECTIVES: Family medicine residents are scored via milestones created by the Accreditation Council for Graduate Medical Education (ACGME) on various clinical domains, including communication. Communication involves a resident's ability to set an agenda, but this is rarely taught in formal education. Our study aimed to examine the relationship between ACGME Milestone achievement and ability to set a visit agenda, as measured by direct observation (DO) forms. METHODS: We examined biannual (December, June) ACGME scores for family medicine residents at an academic institution from 2015-2020. Using faculty DO scores, we rated residents on six items corresponding to agenda setting. We used Spearman and Pearson correlations and two-sample paired t tests to analyze results. RESULTS: We analyzed a total of 246 ACGME scores and 215 DO forms. For first-year residents, we found significant, positive associations between agenda-setting and the total Milestone score (r[190]=.15, P=.034) in December, and in individual (r[190]=.17, P=.020) and total communication scores (r[186]=.16, P=.031), in June. However, for first-year residents, we found no significant correlations with communication scores in December or in the total milestone scores in June. We found significant progression for consecutive years in both communication milestones (t=-15.06, P<0.001) and agenda setting (t=-12.26, P<.001). CONCLUSIONS: The significant associations found in agenda setting with both ACGME total communication and Milestone scores for first-year residents only, suggests that agenda setting may be fundamental in early resident education.

Increased Ileostomy Output as an Indicator of Thyroid Storm in a Patient without an Established History of Underlying Thyroid Disease.

BACKGROUND Thyroid storm, also known as thyrotoxic crisis, is a rare but life-threatening endocrine emergency that presents with multisystem involvement. Patients present with pronounced signs of hyperthyroidism, fever, tachycardia, and differing severities of multisystem dysfunction and decompensation. Early recognition and prompt initiation of treatment are important. The development of thyroid storm in patients with no established history of underlying hyperthyroidism is rare. CASE REPORT In this case report, we describe the occurrence of thyroid storm in a 27-year-old man without an established history of underlying thyroid disease, who was admitted to the Intensive Care Unit (ICU) with a high ileostomy output and fever. Although initially treated for possible sepsis, the diagnosis of thyroid storm was made only after a thorough workup was initiated and he was found to have underlying Graves' disease. Prompt treatment resulted in the resolution of symptoms and avoided potential morbidity and mortality. CONCLUSIONS This case highlights the potential difficulty in diagnosing thyroid storm in a patient admitted to the ICU without an established history of hyperthyroidism. Upgrade in care, timely diagnosis, and initiation of appropriate therapy led to a favorable outcome. Clinicians should consider hyperthyroidism as a possible cause of high ileostomy output, especially when it does not resolve with traditional treatment and no obvious cause can be identified. This case demonstrates the challenges presented when the patient's history and clinical signs are ambiguous and stresses the importance of "outside the box" thinking.

A Case of Lichen Planus Masquerading as Guttate Psoriasis.

Lichen planus (LP) is a chronic inflammatory disease that affects the skin, hair, nails, and mucous membranes, with variants such as drug-induced lichen planus, which is triggered by medications such as angiotensin-converting enzyme (ACE) inhibitors and antimalarials. Guttate psoriasis (GP), a clinical variant of psoriasis, is associated with streptococcal infections and presents with drop-like papules on the trunk and proximal extremities. In this report, we present a case of LP in an atypical location masquerading as GP and the importance of prompt dermatological referral to improve the patient's quality of life. Coexistence and similarities between several variants of LP and plaque psoriasis have been seen in the literature. However, to our knowledge, our report is the first to show LP specifically mimicking GP.

Curation and annotation of planarian gene expression patterns with segmented reference morphologies.

MOTIVATION: Morphological and genetic spatial data from functional experiments based on genetic, surgical and pharmacological perturbations are being produced at an extraordinary pace in developmental and regenerative biology. However, our ability to extract knowledge from these large datasets are hindered due to the lack of formalization methods and tools able to unambiguously describe, centralize and interpret them. Formalizing spatial phenotypes and gene expression patterns is especially challenging in organisms with highly variable morphologies such as planarian worms, which due to their extraordinary regenerative capability can experimentally result in phenotypes with almost any combination of body regions or parts. RESULTS: Here, we present a computational methodology and mathematical formalism to encode and curate the morphological outcomes and gene expression patterns in planaria. Worm morphologies are encoded with mathematical graphs based on anatomical ontology terms to automatically generate reference morphologies. Gene expression patterns are registered to these standard reference morphologies, which can then be annotated automatically with anatomical ontology terms by analyzing the spatial expression patterns and their textual descriptions. This methodology enables the curation and annotation of complex experimental morphologies together with their gene expression patterns in a centralized standardized dataset, paving the way for the extraction of knowledge and reverse-engineering of the much sought-after mechanistic models in planaria and other regenerative organisms. AVAILABILITY AND IMPLEMENTATION: We implemented this methodology in a user-friendly graphical software tool, PlanGexQ, freely available together with the data in the manuscript at https://lobolab.umbc.edu/plangexq. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Scientometric Analysis: University Hospital Versus University College of Medicine.

Many medical specialties use scientometrics to assess the impact of publications, journals, and authors. The aim of this study was to analyze and compare trends of publications from a hospital medical center to publications from a college of medicine connected to that hospital and compare collaboration rates between them to other domestic and international institutions. We used Elsevier's SCOPUS database to compare Penn State College of Medicine (PSCOM) publications to Hershey Medical Center (HMC) publications, analyzing 31,856 total publications. We hypothesized that HMC and PSCOM have room to improve on both internal and international collaborations. Our results show that despite PSCOM's international collaboration being nearly three times higher than HMC, overall international collaboration is less than 2%, far below the US national average.