A Novel Machine Learning Algorithm Combined With Multivariate Analysis for the Prognosis of Renal Collecting Duct Carcinoma

ObjectivesTo investigate the clinical and non-clinical characteristics that may affect the prognosis of patients with renal collecting duct carcinoma (CDC) and to develop an accurate prognostic model for this disease.MethodsThe characteristics of 215 CDC patients were obtained from the U.S. National Cancer Institute’s surveillance, epidemiology and end results database from 2004 to 2016. Univariate Cox proportional hazard model and Kaplan-Meier analysis were used to compare the impact of different factors on overall survival (OS). 10 variables were included to establish a machine learning (ML) model. Model performance was evaluated by the receiver operating characteristic curves (ROC) and calibration plots for predictive accuracy and decision curve analysis (DCA) were obtained to estimate its clinical benefits.ResultsThe median follow-up and survival time was 16 months during which 164 (76.3%) patients died. 4.2, 32.1, 50.7 and 13.0% of patients were histological grade I, II, III, and IV, respectively. At diagnosis up to 61.9% of patients presented with a pT3 stage or higher tumor, and 36.7% of CDC patients had metastatic disease. 10 most clinical and non-clinical factors including M stage, tumor size, T stage, histological grade, N stage, radiotherapy, chemotherapy, age at diagnosis, surgery and the geographical region where the care delivered was either purchased or referred and these were allocated 95, 82, 78, 72, 49, 38, 36, 35, 28 and 21 points, respectively. The points were calculated by the XGBoost according to their importance. The XGBoost models showed the best predictive performance compared with other algorithms. DCA showed our models could be used to support clinical decisions in 1-3-year OS models.ConclusionsOur ML models had the highest predictive accuracy and net benefits, which may potentially help clinicians to make clinical decisions and follow-up strategies for patients with CDC. Larger studies are needed to better understand this aggressive tumor.

Measuring and Improving Evidence-Based Patient Care Using a Web-Based Gamified Approach in Primary Care (QualityIQ): Randomized Controlled Trial

Background Unwarranted variability in clinical practice is a challenging problem in practice today, leading to poor outcomes for patients and low-value care for providers, payers, and patients. Objective In this study, we introduced a novel tool, QualityIQ, and determined the extent to which it helps primary care physicians to align care decisions with the latest best practices included in the Merit-Based Incentive Payment System (MIPS). Methods We developed the fully automated QualityIQ patient simulation platform with real-time evidence-based feedback and gamified peer benchmarking. Each case included workup, diagnosis, and management questions with explicit evidence-based scoring criteria. We recruited practicing primary care physicians across the United States into the study via the web and conducted a cross-sectional study of clinical decisions among a national sample of primary care physicians, randomized to continuing medical education (CME) and non-CME study arms. Physicians “cared” for 8 weekly cases that covered typical primary care scenarios. We measured participation rates, changes in quality scores (including MIPS scores), self-reported practice change, and physician satisfaction with the tool. The primary outcomes for this study were evidence-based care scores within each case, adherence to MIPS measures, and variation in clinical decision-making among the primary care providers caring for the same patient. Results We found strong, scalable engagement with the tool, with 75% of participants (61 non-CME and 59 CME) completing at least 6 of 8 total cases. We saw significant improvement in evidence-based clinical decisions across multiple conditions, such as diabetes (+8.3%, P<.001) and osteoarthritis (+7.6%, P=.003) and with MIPS-related quality measures, such as diabetes eye examinations (+22%, P<.001), depression screening (+11%, P<.001), and asthma medications (+33%, P<.001). Although the CME availability did not increase enrollment in the study, participants who were offered CME credits were more likely to complete at least 6 of the 8 cases. Conclusions Although CME availability did not prove to be important, the short, clinically detailed case simulations with real-time feedback and gamified peer benchmarking did lead to significant improvements in evidence-based care decisions among all practicing physicians. Trial Registration ClinicalTrials.gov NCT03800901; https://clinicaltrials.gov/ct2/show/NCT03800901

Wearable technology to inform the prediction and diagnosis of cardiorespiratory events: a scoping review

Background The need for health systems that allow for continuous monitoring and early adverse event detection in individuals outside of the acute care setting has been highlighted by the global rise in chronic cardiorespiratory diseases and the recent COVID-19 pandemic. Currently, it is unclear what type of evidence exists concerning the use of physiological data collected from commercially available wrist and textile wearables to assist in clinical decision making. The aim of this review was therefore to systematically map and summarize the scientific literature surrounding the use of these wearables in clinical decision making as well as identify knowledge gaps to inform further research. Methodology Six electronic bibliographic databases were systematically searched (Ovid MEDLINE, EMBASE, CINAHL, PubMed, Scopus, and SportsDiscus). Publications from database inception to May 6, 2020 were reviewed for inclusion. Non-indexed literature relevant to this review was also searched systematically. Results were then collated, summarized and reported. Results A total of 107 citations were retrieved and assessed for eligibility with 31 citations included in the final analysis. A review of the 31 papers revealed three major study designs which included (1) observational studies (n = 19), (2) case control series and reports (n = 8), and (3) reviews (n = 2). All papers examined the use of wearable monitoring devices for clinical decisions in the cardiovascular domain, with cardiac arrhythmias being the most studied. When compared to electrocardiogram (ECG) the performance of the wearables in facilitating clinical decisions varied depending upon the type of wearable, user’s activity levels and setting in which they were employed. Observational studies collecting data in the inpatient and outpatient settings were equally represented. Eight case control series and reports were identified which reported on the use of wrist wearables in patients presenting to an emergency department or clinic to aid in the clinical diagnosis of a cardiovascular event. Two narrative reviews were identified which examined the impact of wearable devices in monitoring cardiovascular disease as well as potential challenges they may pose in the future. Conclusions To date, studies employing wearables to facilitate clinical decisions have largely focused upon the cardiovascular domain. Despite the ability of some wearables to collect physiological data accurately, there remains a need for a specialist physician to retrospectively review the raw data to make a definitive diagnosis. Analysis of the results has also highlighted gaps in the literature such as the absence of studies employing wearables to facilitate clinical decisions in the respiratory domain. The disproportionate study of wearables in atrial fibrillation detection in comparison to other cardiac arrhythmias and conditions, as well as the lack of diversity in the sample populations used prevents the generalizability of results.

A quantum-enhanced precision medicine application to support data-driven clinical decisions for the personalized treatment of advanced knee osteoarthritis: development and preliminary validation of precisionKNEE QNN

Background Quantum computing (QC) and quantum machine learning (QML) are promising experimental technologies which can improve precision medicine applications by reducing the computational complexity of algorithms driven by big, unstructured, real-world data. The clinical problem of knee osteoarthritis is that, although some novel therapies are safe and effective, the response is variable, and defining the characteristics of an individual who will respond remains a challenge. In this paper we tested a quantum neural network (QNN) application to support precision data-driven clinical decisions to select personalized treatments for advanced knee osteoarthritis. Methods Following patients consent and Research Ethics Committee approval, we collected clinico-demographic data before and after the treatment from 170 patients eligible for knee arthroplasty (Kellgren-Lawrence grade ≥ 3, OKS ≤ 27, Age ≥ 64 and idiopathic aetiology of arthritis) treated over a 2 year period with a single injection of microfragmented fat. Gender classes were balanced (76 M, 94 F) to mitigate gender bias. A patient with an improvement ≥ 7 OKS has been considered a Responder. We trained our QNN Classifier on a randomly selected training subset of 113 patients to classify responders from non-responders (73 R, 40 NR) in pain and function at 1 year. Outliers were hidden from the training dataset but not from the validation set. Results We tested our QNN Classifier on a randomly selected test subset of 57 patients (34 R, 23 NR) including outliers. The No Information Rate was equal to 0.59. Our application correctly classified 28 Responders out of 34 and 6 non-Responders out of 23 (Sensitivity = 0.82, Specificity = 0.26, F1 Statistic= 0.71). The Positive (LR+) and Negative (LR-) Likelihood Ratios were respectively 1.11 and 0.68. The Diagnostic Odds Ratio (DOR) was equal to 2. Conclusions Preliminary results on a small validation dataset show that quantum machine learning applied to data-driven clinical decisions for the personalized treatment of advanced knee osteoarthritis is a promising technology to reduce computational complexity and improve prognostic performance. Our results need further research validation with larger, real-world unstructured datasets, and clinical validation with an AI Clinical Trial to test model efficacy, safety, clinical significance and relevance at a public health level.

Quantum deep reinforcement learning for clinical decision support in oncology: application to adaptive radiotherapy

Subtle differences in a patient’s genetics and physiology may alter radiotherapy (RT) treatment responses, motivating the need for a more personalized treatment plan. Accordingly, we have developed a novel quantum deep reinforcement learning (qDRL) framework for clinical decision support that can estimate an individual patient’s dose response mid-treatment and recommend an optimal dose adjustment. Our framework considers patients’ specific information including biological, physical, genetic, clinical, and dosimetric factors. Recognizing that physicians must make decisions amidst uncertainty in RT treatment outcomes, we employed indeterministic quantum states to represent human decision making in a real-life scenario. We paired quantum decision states with a model-based deep q-learning algorithm to optimize the clinical decision-making process in RT. We trained our proposed qDRL framework on an institutional dataset of 67 stage III non-small cell lung cancer (NSCLC) patients treated on prospective adaptive protocols and independently validated our framework in an external multi-institutional dataset of 174 NSCLC patients. For a comprehensive evaluation, we compared three frameworks: DRL, qDRL trained in a Qiskit quantum computing simulator, and qDRL trained in an IBM quantum computer. Two metrics were considered to evaluate our framework: (1) similarity score, defined as the root mean square error between retrospective clinical decisions and the AI recommendations, and (2) self-evaluation scheme that compares retrospective clinical decisions and AI recommendations based on the improvement in the observed clinical outcomes. Our analysis shows that our framework, which takes into consideration individual patient dose response in its decision-making, can potentially improve clinical RT decision-making by at least about 10% compared to unaided clinical practice. Further validation of our novel quantitative approach in a prospective study will provide a necessary framework for improving the standard of care in personalized RT.