Early cancer detection is crucial for improved patient outcomes, as evidenced by research on major cancer types emphasizing the impact of timely treatment initiation. This study focuses on identifying pre-diagnosis patterns in prostate cancer, utilizing supervised machine learning to build predictive models analyzing patients' medical activities one year before diagnosis. The dataset, sourced from the All of Us Research Program, specifically targets prostate cancer cases diagnosed between 2010 and 2019. By grouping CPT4 codes in clinically significant categories and employing the XGBoost model in machine learning, the study achieved superior performance with accuracy and area under the curve (AUC) of 0.94 for predicting cancer one month prior to diagnosis and 0.76 five months before diagnosis. In addition, the top important features derived from the model were surgical pathology procedure, cardiac stress tests, hospital inpatient, number of total visits, and diagnostic ultrasound of the head and neck. Despite a decline in accuracy when predicting 5 months and 1 year ahead, this research lays the groundwork for personalized and timely interventions, advancing cancer diagnostics and early intervention strategies.

The diagnostic discrepancies between emergency department diagnosis and hospital discharge diagnosis could affect patient outcomes. The aim of this study was to analyze the diagnostic discrepancies between patient diagnoses in the emergency department and their diagnoses at hospital discharge. The study involved adult patients with Chronic Obstructive Pulmonary Disease in 2017. The diagnostic discrepancies were assessed by evaluating the match percentage between emergency department diagnoses and corresponding hospital discharge diagnoses. The diagnostic mismatch was found in 20% of visits indicating a high prevalence of diagnostic mismatch in the studied population. Future predictive modeling should include clinical notes from the integration of large language models.

The inconsistency in reporting clinical trial outcome measures is a source of variability in clinical trial data which complicates comparability of results across studies. Automatic aggregation of semantically similar outcomes will greatly facilitate the harmonization, cross-trial comparison, and metadata analysis. Interpretation, comparison, and combination of results of clinical trials are reliant on the consistency of outcome reporting. This study aimed to align outcome measures by leveraging the significant capability of large language models. This study proposes a time-efficient, fully automated pipeline to find trials reporting semantically similar outcomes in order to facilitate cross-compatibility and metadata analysis.

Image-based photoplethysmography (iPPG) represents a non-invasive and contactless optical technique extensively utilized for heart rate monitoring. In this methodology, the skin's surface is illuminated by a light source, and a photodetector measures volumetric variations in blood circulation. To circumvent the blinding effects associated with high-intensity visible light sources, this pilot study explores the feasibility of employing infrared (IR) light and a high-speed camera for contactless continuous pulse rate measurement. The IR light is directed onto the facial region, and a high-speed camera, capable of capturing video frames at high sampling rate, records the reflected light, enabling the extraction of iPPG signals from the acquired video frames. The efficacy of the developed algorithm was evaluated in ten experiments, encompassing both resting and exercising conditions. A wrist-worn pulse oximeter served as a reference for comparison. The article details the process of image extraction and signal processing implementation, presenting the outcomes of a signal processing pipeline. Furthermore, the detected pulse rates were compared with measurements obtained from the pulse oximeter.

This pilot study used Long Short-Term Memory (LSTM) and gated recurrent unit (GRU) Recurrent Neural Network (RNN) models to predict exercise exertion levels based on physiological data acquired from wearable devices. The data, including revolutions per minute (RPM), level of oxygen saturation, real-time ECG, and pulse rate were collected during a 16-minute cycling exercise across three intensity levels. Ratings of perceived exertion (RPE) from study subjects were recorded at the end of each consecutive minute during every session. The 16-minute sessions were segmented into eight 2-minute windows, and each window was categorized as either "high exertion" or "low exertion" based on self-reported RPEs. Predictive features were calculated by averaging oxygen saturation levels, heart rate, and RPMs for each 2-minute window. Additionally, heart rate variability (HRV) features were extracted from collected ECG data in both temporal and frequency domains. The selection of predictive features was performed using the minimum redundancy maximum relevance (mRMR) algorithm. The top features selected were employed to train and test the LSTM and the GRU classifiers for predicting the exertion level of the subsequent window. The resulting classifiers exhibited testing accuracies of 78.6% and F-1 scores of 80%. This pilot study underscores the potential of employing a deep learning model for real-time prediction of perceived exercise exertion.

Patients with Chronic Obstructive Pulmonary Disease (COPD) frequently have other comorbidities such as congestive heart failure, hypertension, coronary artery disease, or atrial fibrillation. These conditions exhibit an overlapping set of symptoms which complicates early identification of the primary cause of an acute exacerbation upon admission to a hospital. Timely identification of the underlying condition that led to an acute exacerbation allows expedite prescription of an optimal care plan and reduce unnecessary medical procedures. The aim of this study was to develop a classification model using data from electronic health records for early identification of COPD exacerbation in newly admitted patients. The study cohort included patients with a prior diagnosis of COPD. Three predictive models have been developed among which random forest showed the best performance and resulted in 73% F1score and accuracy of 80%.

The COVID-19 pandemic has impacted every aspect of health delivery and encouraged to replace in-person clinical visits with telecommunications. By providing wireless communication between embedded electronic devices and sensors, telerehabilitation enables constant monitoring of vital body functions, and tracking of physical activities of a person and aids physical therapy. In this paper, we designed and tested two remotely controlled versions of interactive bike (iBikE) systems which communicate through either Wi-Fi or BLE and give the clinical team the capability to monitor exercise progress in real time using simple graphical representation. We used the same hardware and user interface for both designs. The software uses either Wi-Fi or BLE protocol to connect the iBikE equipment and PC tablet. The bike can be used for upper or lower limb rehabilitation. A customized tablet app was developed to provide user interface between the app and the bike sensors. Both bikes were tested with a single group of nine individuals in two separate sessions. Each individual was asked to hand-cycle for three separate sub-sessions (1 minute each for slow, medium, and fast pace) with one-minute rest. During each sub-session, speed of the bikes was measured continuously using a tachometer, in addition to reading speed values from the iBikE app, to compare the functionality and accuracy of the measured data. Measured RPMs in each sub-session from iBikE and tachometer were further divided into 4 categories: 10-second bins (6 bins), 20-second bins (3 bins), 30-second bins (2 bins), and RPMs in each sub-session (1 minute, 1 bin). Then, the mean difference of each category (iBikE, tachometer) was calculated for each sub-session. Finally, mean and standard deviation (SD) of the calculated mean differences were reported for all individuals. We saw decreasing trend in both mean and SD from 10 second to 1 minute measurement. For BLE iBikE system, minimum mean RPM difference was$0.2 \pm 0.3$in one-minute sub-session with medium speed. This number was$0.21 \pm 0.21$in one-minute sub-session with slow speed for Wi-Fi iBikE system. Thus, testing confirmed high accuracy of our interfaces.

Personalized exercise interventions comprising aerobic, resistance and corrective therapeutic components were shown to result in endurance and strength improvement and reduction of fatigue and spasticity in patients with multiple sclerosis (PwMS). However, telerehabilitation support of multipronged exercise programs in PwMS has received limited attention. The goal of this project was to introduce telerehabilitation support of individualized multipronged exercise programs and to assess feasibility assessment of the initial prototype in PwMS. Feasibility was assessed in 10 PwMS who were asked to perform a standardized set of tasks essential for using the telerehabilitation system independently and score each task. Our results indicated high level of acceptance of the system by these patients. On average, it took about 1-2 minutes for the patients to complete the study tasks essential for system operation. They were able to successfully use the system and follow their individualized exercise prescription. The resulting system will have to undergo a definitive systematic evaluation in a randomized controlled trial to demonstrate its clinical impact.

Medical errors and patient safety have been receiving significant attention since the landmark publication by Institute of Medicine in 2000. However, characteristics of physicians implicated in frequent medical errors were not studied systematically. We used National Practitioner Data Bank (NPDB) containing malpractice claims since 1990 to identify characteristics predictive of physicians with frequent malpractice claims. We separated all malpractice records for US physicians into two groups according to the total number of malpractice records (0: less than 5 records, 1: more than 4 records) and compared characteristics of the first malpractice record in each group. Overall, 137,590 unique records were analyzed. Four percent of physicians (5371) had 5 or more malpractice claims. Bivariate statistics, cross-correlation and principal component analysis were used to identify predictive features. Logistic regression was used for predictive modeling. Resulting model allowed prediction of physicians with frequent malpractice records based on the following characteristics of the first malpractice record: allegation type, practitioner age, number of years from graduation to the first malpractice claim.

Objective: Temporomandibular Disorder (TMD) is a jaw joint dysfunction characterized by various symptoms including pain in the jaw joint, facial structures, head, and muscles controlling the joint movement. In many patients TMD results in chronic pain and other related conditions. As the exact cause and cure of TMD remains unknown, individuals affected by TMD continue to experience high amount of tension and anxiety. With wide accessibility of internet to the public, TMD patients gained new platform for empowerment, social support and informational gain. However, components of empowerment obtained by TMD patients from online support groups were not studied systematically. The aim of this study was to identify major dimensions of empowerment available for users of TMD online groups and to explore how they communicate personal experience. Methods: Qualitative content analysis of highly referenced TMD Online Support Group was carried out using patient empowerment and support themes developed for this study. Results: All major dimensions of empowerment were present in online exchange between TMD patients. The analysis indicated that TMD support group members most frequently discussed emotional issues (46.2%) followed by informational issues (26.2%) with counseling issues (10.8%) and cost of care issues (4.62%) least frequently discussed. Conclusion: The significant proportion of emotional and informational support indicated that online interaction facilitated patient empowerment by providing essential components for TMD self-management that were not made available sufficiently via alternatives venues. Practice implications: People have grown more familiar and comfortable with sharing their personal stories or situations with internet based support groups. Online forums offer effective means of patient empowerment by facilitating continuous engagement of patients with TMD in their self-care.

Polypharmacy has been shown to be a source of multiple adverse drug events. Recent studies demonstrated that pharmacogenomic testing may be instrumental in optimizing medication regimen. However, majority of current pharmacogenomic decision support tools provide assessment only of single drug-gene interactions without taking into account complex drug-drug and drug-drug-gene interactions which are prevalent in people with polypharmacy and were shown to be a significant source of adverse drug events. The main objective of this project is development of comprehensive pharmacogenomic decision support for medication risk assessment in people with polypharmacy. To achieve this goal, the project addressed two aims: (1) development of comprehensive knowledge repository of actionable pharmacogenes; (2) introduction of scoring approaches reflecting potential adverse effect risk levels of complex medication regimens accounting for pharmacogenomic polymorphisms and multiple drug metabolizing pathways. After pharmacogenomic knowledge repository was introduced, additive and multiplicative algorithms for medication risk assessment have been implemented. An initial prototype assessment demonstrated feasibility of our approach and identified steps for improving risk scoring algorithms.

Polypharmacy in older adults has been shown to be a significant risk factor for adverse drug events. Pharmacogenomic testing provides insights regarding individual drug efficacy and may be instrumental in optimizing medication regimen. However, majority of current pharmacogenomic decision support tools provide assessment only of single drug-gene interactions without taking into account complex drug-drug and drug-drug-gene interactions which are prevalent in people with polypharmacy and can result in adverse drug events or insufficient drug efficacy. The main objective of this project was to develop comprehensive pharmacogenomic decision support for medication risk assessment in people with polypharmacy that simultaneously accounts for multiple drug and gene effects. To achieve this goal, the project addressed two aims: (1) development of comprehensive knowledge repository of actionable pharmacogenes; (2) introduction of scoring approaches reflecting potential adverse effect risk levels of complex medication regimens accounting for pharmacogenomic polymorphisms and multiple drug metabolizing pathways. After pharmacogenomic knowledge repository was introduced, a scoring algorithm has been built and pilot-tested using a limited data set. The resulting total risk score for frequently hospitalized older adults with polypharmacy (72.04±17.84) was statistically significantly different (p<;0.05) from the total risk score for older adults with polypharmacy with low hospitalization rate (8.98±2.37). An initial prototype assessment demonstrated feasibility of our approach and identified steps for improving risk scoring algorithms.

Recent studies demonstrated that multipronged exercise interventions comprising aerobic, resistance and corrective therapeutic components resulted in endurance and strength improvement and reduction of fatigue and spasticity in patients with multiple sclerosis (PwMS). However, telerehabilitation support of multipronged exercise programs in PwMS has received limited attention. The goal of this project was to introduce telerehabilitation support of individualized multipronged exercise programs and to assess feasibility assessment of the initial prototype in PwMS. In this project, design of telerehabilitation system supporting individualized multipronged exercise program was introduced and its feasibility was assessed in PwMS. Our results indicated high level of acceptance of the system by these patients. On average, it took about 1-2 minutes for the patients to complete the study tasks essential for system operation. They were able to successfully use the system and follow their individualized exercise prescription. The resulting system will have to undergo a definitive systematic evaluation in randomized clinical trial to demonstrate its clinical impact.

Exercise exertion results in activation of sympathetic nervous system. Heart rate variability (HRV) has been used to analyze activity of sympathetic nervous system (ANS). However, approaches to use HRV for exercise exertion analysis were not explored systematically. The main goal of this study was to develop classification algorithms to determine level of exercise exertion in real time and to compare potential of HRV time domain parameters versus HRV frequency domain parameters versus combined time and frequency parameter set. Discriminant analysis was used to identify optimal parameter sets and to develop algorithms for classification of exercise exertion levels. Time-domain HRV parameters demonstrated higher classification accuracy (95.6%) as compared to frequency-domain parameters (82.2%). Combing HRV parameters from time and frequency domains improves classification accuracy (97.8%). Our results suggested that HRV analysis can be used to automatically classify exercise exertion levels. Future studies should focus on more granular approach in identifying different stages of exercise process. Evaluation of classification algorithms should be based on larger sample of diverse representatives of different age, sex and health condition groups.

Gaming platforms enjoy growing recognition as useful means to facilitate consumer health. The potential of such widely used platforms as Xbox for self-management of chronic health conditions has not yet been fully uncovered. We developed a Home Automated Telemanagement (HAT) system which utilizes Xbox to access personal health record, receive self-care support, and exchange information with health care providers. The system questions patients with congestive heart failure (CHF) to monitor symptoms, weight changes, and quality of life while educating the patient on their disease. The system is designed to run on the Microsoft Xbox videogame console using an active internet connection and a connection to a computer running Windows Media Center. It questions the patient daily on their condition, monitors their weight, and provides the patient with instant feedback on their condition in the form of a 3 zone CHF action plan. The system is designed to be as simple as possible, making it usable by patients with no prior computer or videogame experience. This telemanagement system has been successfully designed and implemented to optimize the care of patients with CHF.

Quasi-experimental study designs, often described as nonrandomized, pre-post intervention studies, are common in the medical informatics literature. Yet little has been written about the benefits and limitations of the quasi-experimental approach as applied to informatics studies. This paper outlines a relative hierarchy and nomenclature of quasi-experimental study designs that is applicable to medical informatics intervention studies. In addition, the authors performed a systematic review of two medical informatics journals, the Journal of the American Medical Informatics Association (JAMIA) and the International Journal of Medical Informatics (IJMI), to determine the number of quasi-experimental studies published and how the studies are classified on the above-mentioned relative hierarchy. They hope that future medical informatics studies will implement higher level quasi-experimental study designs that yield more convincing evidence for causal links between medical informatics interventions and outcomes.

Recent studies demonstrated that successful chronic disease management requires a coordinated effort that includes health care providers who follow recent clinical guidelines in their every day practice, educated patients who are adherent to their prescribed treatment plans and comprehensive patient-provider communication. Studies employing telecommunication technologies have been successful in affecting the major components of chronic disease care including physician practice patterns, patient adherence to therapy, and patient-provider communication. These studies were focused mainly on an improvement of a single component of clinical care process. To date there have been no studies targeting a comprehensive coordinated approach that concurrently employs all of these components in an integrated framework linked to primary care. In this study we propose a multi-component telemanagement system providing an integrated support both to clinicians and patients in implementing hypertension treatment guidelines promulgated by the Seventh Report of the Joint National Committee on Prevention detection, Evaluation, and Treatment of High Blood Pressure.

This project is aimed to assess feasibility of a novel patient-centered education model combining behavioral theories and interactive computer media to promote patient-driven care. The following specific features of the proposed model has been implemented: I) the educational intervention targets specific constructs, shown to facilitate patient self-care, in a unique way which accounts for psycho-social profile, disease history and personal preferences; 2) the educational intervention is linked and coordinated with patient electronic medical record; 3) computer-assisted education is driven by theories of adult learning and could be adapted to different levels of education, cognition or individual disabilities; 4) specific components has been included to achieve sustainability of benefits attained by initial patient education. Feasibility of the model in patients with multiple sclerosis has been evaluated.

The Internet-based home automated telemanagement (HAT) system has been designed to support a multi-disciplinary approach in patient self-management which includes regular patient assessment, disease-specific education, control of patient compliance with treatment plans, implementation of health behavior change models and social support both for patients and caregivers. The goals of this project were: (1) To refine the HAT program to fully implement the multidisciplinary model for telemanagement of COPD patients; (2) To evaluate in a pilot study feasibility and acceptance of HAT by COPD patients. The COPD HAT interface can be operated by a mouse, keyboard or voice. Specific features have been introduced into presentation and navigation of the HAT home unit to accommodate for possible limitations in vision, motion or cognition by the potential HAT users. Feasibility evaluation showed that the COPD patients were able to use the COPD HAT system successfully regardless of their age, education or prior experience in using computers.

The aim of this project was construction of a universal platform for rapid development of interactive computer-based health education programs. Pocket PC was used to pilot test the platform and to implement two educational curricula: for asthma and anticoagulant therapy. Qualitative analysis of the system acceptance in 46 patients showed that computer-assisted education can be successfully implemented on the basis of a handheld computer.