Deep-Learning Approach to Predict Survival Outcomes Using Wearable Actigraphy Device Among End-Stage Cancer Patients.

Survival prediction is highly valued in end-of-life care clinical practice, and patient performance status evaluation stands as a predominant component in survival prognostication. While current performance status evaluation tools are limited to their subjective nature, the advent of wearable technology enables continual recordings of patients' activity and has the potential to measure performance status objectively. We hypothesize that wristband actigraphy monitoring devices can predict in-hospital death of end-stage cancer patients during the time of their hospital admissions. The objective of this study was to train and validate a long short-term memory (LSTM) deep-learning prediction model based on activity data of wearable actigraphy devices. The study recruited 60 end-stage cancer patients in a hospice care unit, with 28 deaths and 32 discharged in stable condition at the end of their hospital stay. The standard Karnofsky Performance Status score had an overall prognostic accuracy of 0.83. The LSTM prediction model based on patients' continual actigraphy monitoring had an overall prognostic accuracy of 0.83. Furthermore, the model performance improved with longer input data length up to 48 h. In conclusion, our research suggests the potential feasibility of wristband actigraphy to predict end-of-life admission outcomes in palliative care for end-stage cancer patients. Clinical Trial Registration: The study protocol was registered on ClinicalTrials.gov (ID: NCT04883879).

Diagnostic Accuracy of Female Pelvic Ultrasonography in Differentiating Precocious Puberty From Premature Thelarche: A Systematic Review and Meta-analysis.

Background: The gonadotropin-releasing hormone (GnRH) stimulation test is the benchmark for diagnosing precocious puberty (PP). However, it is invasive, time-consuming, costly, and may create an unpleasant experience for participants. Moreover, some overlaps may occur between PP and premature thelarche (PT) in the early stage of PP. Female pelvic ultrasonography may provide additional information to help differentiate PP from PT and subsequently initiate early treatment. In this study, we aimed to first directly compare pelvic ultrasonography parameters between PP and PT groups and secondly, investigate their diagnostic accuracy compared with the GnRH stimulation test. Methods: A systematic search of the PubMed/MEDLINE, EMBASE, Scopus, and Cochrane Library databases was performed up to March 31, 2021. All types of studies, except for case reports and review articles, were included. The GnRH stimulation test was used to confirm PP diagnosis. Those whose organic conditions might cause PP were excluded. The mean, standard deviation, sensitivity, and specificity of each parameter were documented. Forest plots were constructed to display the estimated standardized mean differences (SMDs) from each included study and the overall calculations. A bivariate model was used to calculate the pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR). Results: A total of 13 studies were included for analysis. The SMDs (95% confidence interval - CI) in ovarian volume, fundal-cervical ratio, uterine length, uterine cross-sectional area, and uterine volume between PP and PT groups were 1.12 (0.78-1.45; p < 0.01), 0.90 (0.07-1.73; p = 0.03), 1.38 (0.99-1.78; p < 0.01), 1.06 (0.61-1.50; p < 0.01), and 1.21 (0.84-1.58; p <0.01), respectively. A uterine length of 3.20 cm yielded a pooled sensitivity of 81.8% (95% CI 78.3%-84.9%), specificity of 82.0% (95% CI 61.0%-93.0%), PLR of 4.56 (95% CI 2.15-9.69), NLR of 0.26 (95% CI 0.17-0.39), and DOR of 19.62 (95% CI 6.45-59.68). The area under the summary receiver operating characteristics curve was 0.82. Conclusion: Female pelvic ultrasonography may serve as a complementary tool to the GnRH stimulation test in differentiating PP from PT. Systematic Review Registration: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021232427, ID: CRD42021232427.

Diagnostic Accuracy of Ambulatory Devices in Detecting Atrial Fibrillation: Systematic Review and Meta-analysis.

BACKGROUND: Atrial fibrillation (AF) is the most common cardiac arrhythmia worldwide. Early diagnosis of AF is crucial for preventing AF-related morbidity, mortality, and economic burden, yet the detection of the disease remains challenging. The 12-lead electrocardiogram (ECG) is the gold standard for the diagnosis of AF. Because of technological advances, ambulatory devices may serve as convenient screening tools for AF. OBJECTIVE: The objective of this review was to investigate the diagnostic accuracy of 2 relatively new technologies used in ambulatory devices, non-12-lead ECG and photoplethysmography (PPG), in detecting AF. We performed a meta-analysis to evaluate the diagnostic accuracy of non-12-lead ECG and PPG compared to the reference standard, 12-lead ECG. We also conducted a subgroup analysis to assess the impact of study design and participant recruitment on diagnostic accuracy. METHODS: This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. MEDLINE and EMBASE were systematically searched for articles published from January 1, 2015 to January 23, 2021. A bivariate model was used to pool estimates of sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and area under the summary receiver operating curve (SROC) as the main diagnostic measures. Study quality was evaluated using the quality assessment of diagnostic accuracy studies (QUADAS-2) tool. RESULTS: Our search resulted in 16 studies using either non-12-lead ECG or PPG for detecting AF, comprising 3217 participants and 7623 assessments. The pooled estimates of sensitivity, specificity, PLR, NLR, and diagnostic odds ratio for the detection of AF were 89.7% (95% CI 83.2%-93.9%), 95.7% (95% CI 92.0%-97.7%), 20.64 (95% CI 10.10-42.15), 0.11 (95% CI 0.06-0.19), and 224.75 (95% CI 70.10-720.56), respectively, for the automatic interpretation of non-12-lead ECG measurements and 94.7% (95% CI 93.3%-95.8%), 97.6% (95% CI 94.5%-99.0%), 35.51 (95% CI 18.19-69.31), 0.05 (95% CI 0.04-0.07), and 730.79 (95% CI 309.33-1726.49), respectively, for the automatic interpretation of PPG measurements. CONCLUSIONS: Both non-12-lead ECG and PPG offered high diagnostic accuracies for AF. Detection employing automatic analysis techniques may serve as a useful preliminary screening tool before administering a gold standard test, which generally requires competent physician analyses. Subgroup analysis indicated variations of sensitivity and specificity between studies that recruited low-risk and high-risk populations, warranting future validity tests in the general population. TRIAL REGISTRATION: PROSPERO International Prospective Register of Systematic Reviews CRD42020179937; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=179937.