Search

Objective: The aim of this study was to evaluate the accuracy, comprehensiveness, and safety of a publicly available large language model (LLM)-ChatGPT in the sub-domain of glaucoma., Design: Evaluation of diagnostic test or technology., Subjects Participants And/or Controls: We seek to evaluate the responses of an artificial intelligence chatbot ChatGPT (version GPT-3.5, OpenAI)., Methods Intervention or Testing: We curated 24 clinically relevant questions in the domain of glaucoma. The questions spanned four categories: pertaining to diagnosis, treatment, surgeries, and ocular emergencies. Each question was posed to the LLM and the responses obtained were graded by an expert grader panel of three glaucoma specialists with combined experience of more than 30 years in the field. For responses which performed poorly, the LLM was further prompted to self-correct. The subsequent responses were then re-evaluated by the expert panel., Main Outcome Measures: Accuracy, comprehensiveness, and safety of the responses of a public domain LLM., Results: There were a total of 24 questions and three expert graders with a total number of responses of n  = 72. The scores were ranked from 1 to 4, where 4 represents the best score with a complete and accurate response. The mean score of the expert panel was 3.29 with a standard deviation of 0.484. Out of the 24 question-response pairs, seven (29.2%) of them had a mean inter-grader score of 3 or less. The mean score of the original seven question-response pairs was 2.96 which rose to 3.58 after an opportunity to self-correct (z-score - 3.27, p  = 0.001, Mann-Whitney U). The seven out of 24 question-response pairs which performed poorly were given a chance to self-correct. After self-correction, the proportion of responses obtaining a full score increased from 22/72 (30.6%) to 12/21 (57.1%), ( p  = 0.026, χ 2 test)., Conclusion: LLMs show great promise in the realm of glaucoma with additional capabilities of self-correction. The application of LLMs in glaucoma is still in its infancy, and still requires further research and validation., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The author(s) declared that they were an editorial board member of Frontiers, at the time of submission. This had no impact on the peer review process and the final decision., (Copyright © 2024 Tan, Tham, Koh, Loon, Aquino, Lun, Cheng, Ngiam and Tan.)

Improved upstream primary prevention of cardiovascular disease (CVD) would enable more individuals to lead lives free of CVD. However, there remain limitations in the current provision of CVD primary prevention, where artificial intelligence (AI) may help to fill the gaps. Using the data informatics capabilities at the National University Health System (NUHS), Singapore, empowered by the Endeavour AI system, and combined large language model (LLM) tools, our team has created a real-time dashboard able to capture and showcase information on cardiovascular risk factors at both individual and geographical level- CardioSight. Further insights such as medication records and data on area-level socioeconomic determinants allow a whole-of-systems approach to promote healthcare delivery, while also allowing for outcomes to be tracked effectively. These are paired with interventions, such as the CHronic diseAse Management Program (CHAMP), to coordinate preventive cardiology care at a pilot stage within our university health system. AI tools in synergy allow the identification of at-risk patients and actionable steps to mitigate their health risks, thereby closing the gap between risk identification and effective patient care management in a novel CVD prevention workflow., Competing Interests: The authors declare no conflicts of interest relevant to the publication of this study., (© 2024 The Authors.)

Background: A plethora of weight management apps are available, but many individuals, especially those living with overweight and obesity, still struggle to achieve adequate weight loss. An emerging area in weight management is the support for one's self-regulation over momentary eating impulses., Objective: This study aims to examine the feasibility and effectiveness of a novel artificial intelligence-assisted weight management app in improving eating behaviors in a Southeast Asian cohort., Methods: A single-group pretest-posttest study was conducted. Participants completed the 1-week run-in period of a 12-week app-based weight management program called the Eating Trigger-Response Inhibition Program (eTRIP). This self-monitoring system was built upon 3 main components, namely, (1) chatbot-based check-ins on eating lapse triggers, (2) food-based computer vision image recognition (system built based on local food items), and (3) automated time-based nudges and meal stopwatch. At every mealtime, participants were prompted to take a picture of their food items, which were identified by a computer vision image recognition technology, thereby triggering a set of chatbot-initiated questions on eating triggers such as who the users were eating with. Paired 2-sided t tests were used to compare the differences in the psychobehavioral constructs before and after the 7-day program, including overeating habits, snacking habits, consideration of future consequences, self-regulation of eating behaviors, anxiety, depression, and physical activity. Qualitative feedback were analyzed by content analysis according to 4 steps, namely, decontextualization, recontextualization, categorization, and compilation., Results: The mean age, self-reported BMI, and waist circumference of the participants were 31.25 (SD 9.98) years, 28.86 (SD 7.02) kg/m 2 , and 92.60 (SD 18.24) cm, respectively. There were significant improvements in all the 7 psychobehavioral constructs, except for anxiety. After adjusting for multiple comparisons, statistically significant improvements were found for overeating habits (mean -0.32, SD 1.16; P<.001), snacking habits (mean -0.22, SD 1.12; P<.002), self-regulation of eating behavior (mean 0.08, SD 0.49; P=.007), depression (mean -0.12, SD 0.74; P=.007), and physical activity (mean 1288.60, SD 3055.20 metabolic equivalent task-min/day; P<.001). Forty-one participants reported skipping at least 1 meal (ie, breakfast, lunch, or dinner), summing to 578 (67.1%) of the 862 meals skipped. Of the 230 participants, 80 (34.8%) provided textual feedback that indicated satisfactory user experience with eTRIP. Four themes emerged, namely, (1) becoming more mindful of self-monitoring, (2) personalized reminders with prompts and chatbot, (3) food logging with image recognition, and (4) engaging with a simple, easy, and appealing user interface. The attrition rate was 8.4% (21/251)., Conclusions: eTRIP is a feasible and effective weight management program to be tested in a larger population for its effectiveness and sustainability as a personalized weight management program for people with overweight and obesity., Trial Registration: ClinicalTrials.gov NCT04833803; https://classic.clinicaltrials.gov/ct2/show/NCT04833803., (©Han Shi Jocelyn Chew, Nicholas WS Chew, Shaun Seh Ern Loong, Su Lin Lim, Wai San Wilson Tam, Yip Han Chin, Ariana M Chao, Georgios K Dimitriadish, Yujia Gao, Jimmy Bok Yan So, Asim Shabbir, Kee Yuan Ngiam. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 07.05.2024.)

Background: Most thyroid nodules are benign. It is important to determine the likelihood of malignancy in such nodules to avoid unnecessary surgery. The primary objective of this study was to characterize the genetic landscape and the performance of a multigene genomic classifier in fine-needle aspiration (FNA) biopsies of cytologically indeterminate thyroid nodules in a Southeast Asian cohort. The secondary objective was to assess the predictive contribution of clinical characteristics to thyroid malignancy., Methods: This prospective, multicenter, blinded study included 132 patients with 134 nodules. Molecular testing (MT) with ThyroSeq v3 was performed on clinical or ex-vivo FNA samples. Centralized pathology review also was performed., Results: Of 134 nodules, consisting of 61% Bethesda category III, 20% category IV, and 19% category V cytology, and 56% were histologically malignant. ThyroSeq yielded negative results in 37.3% of all FNA samples and in 42% of Bethesda category III-IV cytology nodules. Most positive samples had RAS-like (41.7%), followed by BRAF-like (22.6%), and high-risk (17.9%) alterations. Compared with North American patients, the authors observed a higher proportion of RAS-like mutations, specifically NRAS, in Bethesda categories III and IV and more BRAF-like mutations in Bethesda category III. The test had sensitivity, specificity, negative predictive value, and positive predictive value of 89.6%, 73.7%, 84.0%, and 82.1%, respectively. The risk of malignancy was predicted by positive MT and high-suspicion ultrasound characteristics according to American Thyroid Association criteria., Conclusions: Even in the current Southeast Asian cohort with nodules that had a high pretest cancer probability, MT could lead to potential avoidance of diagnostic surgery in 42% of patients with Bethesda category III-IV nodules. MT positivity was a stronger predictor of malignancy than clinical parameters., (© 2024 The Authors. Cancer Cytopathology published by Wiley Periodicals LLC on behalf of American Cancer Society.)

Introduction: With in increase in interest to incorporate artificial intelligence (AI) into weight management programs, we aimed to examine user perceptions of AI-based mobile apps for weight management in adults with overweight and obesity., Methods: 280 participants were recruited between May and November 2022. Participants completed a questionnaire on sociodemographic profiles, Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), and Self-Regulation of Eating Behavior Questionnaire. Structural equation modeling was performed using R. Model fit was tested using maximum-likelihood generalized unweighted least squares. Associations between influencing factors were analyzed using correlation and linear regression., Results: 271 participant responses were analyzed, representing participants with a mean age of 31.56 ± 10.75 years, median (interquartile range) BMI, and waist circumference of 27.2 kg/m 2 (24.2-28.4 kg/m 2 ) and 86.4 (80.0-94.0) cm, respectively. In total, 188 (69.4%) participants intended to use AI-assisted weight loss apps. UTAUT2 explained 63.3% of the variance in our intention of the sample to use AI-assisted weight management apps with satisfactory model fit: CMIN/df = 1.932, GFI = 0.966, AGFI = 0.954, NFI = 0.909, CFI = 0.954, RMSEA = 0.059, SRMR = 0.050. Only performance expectancy, hedonic motivation, and the habit of using AI-assisted apps were significant predictors of intention. Comparison with existing literature revealed vast variabilities in the determinants of AI- and non-AI weight loss app acceptability in adults with and without overweight and obesity. UTAUT2 produced a good fit in explaining the acceptability of AI-assisted apps among a multi-ethnic, developed, southeast Asian sample with overweight and obesity., Conclusion: UTAUT2 model is recommended to guide the development of AI-assisted weight management apps among people with overweight and obesity., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Chew, Achananuparp, Dalakoti, Chew, Chin, Gao, So, Shabbir, Peng and Ngiam.)

Current and future healthcare professionals are generally not trained to cope with the proliferation of artificial intelligence (AI) technology in healthcare. To design a curriculum that caters to variable baseline knowledge and skills, clinicians may be conceptualized as "consumers", "translators", or "developers". The changes required of medical education because of AI innovation are linked to those brought about by evidence-based medicine (EBM). We outline a core curriculum for AI education of future consumers, translators, and developers, emphasizing the links between AI and EBM, with suggestions for how teaching may be integrated into existing curricula. We consider the key barriers to implementation of AI in the medical curriculum: time, resources, variable interest, and knowledge retention. By improving AI literacy rates and fostering a translator- and developer-enriched workforce, innovation may be accelerated for the benefit of patients and practitioners., Competing Interests: Declaration of interests D.H. is a co-inventor listed in current and pending patents pertaining to AI-based personalized medicine. D.S.W.T. and T.Y.W. are the co-inventors of a deep learning system for retinal diseases., (Copyright © 2023 The Author(s). Published by Elsevier Inc. All rights reserved.)

Background: Successful implementation of artificial intelligence in gastroenterology and hepatology practice requires more than technology. There are ethical, legal, and social issues that need to be settled., Aim: A group consisting of AI developers (engineer), AI users (gastroenterologist, hepatologist, and surgeon) and AI regulators (ethicist and administrator) formed a Working Group to draft these Positions Statements with the objective of arousing public and professional interest and dialogue, to promote ethical considerations when implementing AI technology, to suggest to policy makers and health authorities relevant factors to take into account when approving and regulating the use of AI tools, and to engage the profession in preparing for change in clinical practice., Statements: These series of Position Statements point out the salient issues to maintain the trust between care provider and care receivers, and to legitimize the use of a non-human tool in healthcare delivery. It is based on fundamental principles such as respect, autonomy, privacy, responsibility, and justice. Enforcing the use of AI without considering these factor risk damaging the doctor-patient relationship., (© 2023 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Purpose: To evaluate the operative duration and clinical performance of ophthalmology residents performing standard phacoemulsification cataract surgeries using information available from electronic health records (EHR)., Methods: This is a retrospective cohort study. De-identified surgical records of all standard phacoemulsifications performed in a tertiary institution between 1st January 2015 and 8th August 2018 were retrieved from the hospital EHR. The main outcome measures were improvement in operative duration with case experience, corrected distance visual acuity (CDVA) improvement, and intra-operative complication rates., Results: Twelve ophthalmology residents performed a total of 1427 standard phacoemulsifications. The median operative duration was 27 min (interquartile range, 22-34 min), which improved from 31 to 24 min (before the 101st case [Group 1] versus 101st case onwards [Group 2], p < 0.001). Gradient change analysis (non-linear regression) showed significant reduction until the 100th case (p = 0.043). Older patients (0.019), worse pre-operative CDVA (0.343), and surgery performed by Group 1 (1.115) were significantly associated with operative duration above 30 min. LogMAR CDVA improved from a mean of 0.57 ± 0.52 pre-operatively to 0.10 ± 0.18 post-operatively (p < 0.001). Posterior capsule rupture (PCR) rate decreased from 4.0% [Group 1] to 2.1% [Group 2] (p = 0.096), while overall complication rate decreased from 8.9% to 3.1% (p < 0.001)., Conclusion: The median operative duration reduced consistently with surgical experience for the first 100 cases. Older patients, poorer pre-operative VA, and surgical experience of less than 100 cases were significantly associated with an operative duration above 30 min. There was a statistically significant decrease in complication rate between Group 1 and 2., (© 2023. The Author(s), under exclusive licence to Springer Nature B.V.)

Objective: Thyroid nodules are extremely common, with prevalence rate up to 68%, yet only 7-15% of these are malignant. Many nodules require surveillance and 2-dimensional ultrasound (2D US) is used. Issues include the huge workload of obtaining and labeling images and difficulty comparing sizes of nodules over time due to large inter-operator variability. Inaccuracies may result in unnecessary FNAC or missed diagnosis of malignant nodules., Methods: We compared two techniques: freehand plain 2D US against freehand 2D US with gyroscopic guidance, both followed by 3D reconstruction using software. We measured the volume of nodules and a normal thyroid gland., Results: We found 2D US with gyroscopic guidance to be superior to plain 2D US as 3D reconstructions of greater accuracy are produced. The volume of the thyroid lobe measured 8.42 cm 3  ± 0.94 was reasonably close to the normal average volume. However, the measured volume of the ellipsoidal nodule by the software is 8.69 cm 3  ± 0.97 while the measured volume of the spherical nodule is 7.09 cm 3  ± 0.79. As the expected volume of the nodules were 4.24cm 3 and 4.19 cm 3 respectively, the measured volume of the nodule was not accurate. The time taken to characterise nodules was reduced greatly from over 30 min in usual procedure to less than 10 min., Conclusion: We find 3D US promising for evaluating size of thyroid nodules, with potential to study other TIRAD characteristics. Freehand 2D US with gyroscopic guidance shows the most promise for producing reliable, accurate and faster 3D reconstructions of thyroid nodules., (© 2022. Società Italiana di Ultrasonologia in Medicina e Biologia (SIUMB).)

Automatic coronary artery stenosis grading plays an important role in the diagnosis of coronary artery disease. Due to the difficulty of learning the informative features from varying grades of stenosis, it is still a challenging task to identify coronary artery stenosis from coronary CT angiography (CCTA). In this paper, we propose a context-aware deep network (CADN) for coronary artery stenosis classification. The proposed method integrates 3D CNN with Transformer to improve the feature representation of coronary artery stenosis in CCTA. We evaluate the proposed method on a multicenter dataset (APOLLO study with NCT05509010). Experimental results show that our proposed method can achieve the accuracy of 0.84, 0.83, and 0.86 for stenosis diagnosis on the lesion, artery, and patient levels, respectively.

Artificial intelligence (AI) and digital innovation are transforming healthcare. Technologies such as machine learning in image analysis, natural language processing in medical chatbots and electronic medical record extraction have the potential to improve screening, diagnostics and prognostication, leading to precision medicine and preventive health. However, it is crucial to ensure that AI research is conducted with scientific rigour to facilitate clinical implementation. Therefore, reporting guidelines have been developed to standardise and streamline the development and validation of AI technologies in health. This commentary proposes a structured approach to utilise these reporting guidelines for the translation of promising AI techniques from research and development into clinical translation, and eventual widespread implementation from bench to bedside., Competing Interests: Dr Daniel SW Ting holds a patent on a deep learning system for detection of retinal diseases, co-founded and holds equity of EyRIS Singapore. Dr Carolyn SP Lam holds a patent on a deep learning system for detection of cardiac disease, co-founded and holds equity in Us2.ai. Dr Dean Ho is scientific co-founder and shareholder of KYAN Therapeutics. He is also a co-inventor of pending patents pertaining to AI-based drug development and personalised medicine.

(1) Background: Intra-operative neuronavigation is currently an essential component to most neurosurgical operations. Recent progress in mixed reality (MR) technology has attempted to overcome the disadvantages of the neuronavigation systems. We present our experience using the HoloLens 2 in neuro-oncology for both intra- and extra-axial tumours. (2) Results: We describe our experience with three patients who underwent tumour resection. We evaluated surgeon experience, accuracy of superimposed 3D image in tumour localisation with standard neuronavigation both pre- and intra-operatively. Surgeon training and usage for HoloLens 2 was short and easy. The process of image overlay was relatively straightforward for the three cases. Registration in prone position with a conventional neuronavigation system is often difficult, which was easily overcome during use of HoloLens 2. (3) Conclusion: Although certain limitations were identified, the authors feel that this system is a feasible alternative device for intra-operative visualization of neurosurgical pathology. Further studies are being planned to assess its accuracy and suitability across various surgical disciplines.

Advancements in high-throughput sequencing have yielded vast amounts of genomic data, which are studied using genome-wide association study (GWAS)/phenome-wide association study (PheWAS) methods to identify associations between the genotype and phenotype. The associated findings have contributed to pharmacogenomics and improved clinical decision support at the point of care in many healthcare systems. However, the accumulation of genomic data from sequencing and clinical data from electronic health records (EHRs) poses significant challenges for data scientists. Following the rise of artificial intelligence (AI) technology such as machine learning and deep learning, an increasing number of GWAS/PheWAS studies have successfully leveraged this technology to overcome the aforementioned challenges. In this review, we focus on the application of data science and AI technology in three areas, including risk prediction and identification of causal single-nucleotide polymorphisms, EHR-based phenotyping and CRISPR guide RNA design. Additionally, we highlight a few emerging AI technologies, such as transfer learning and multi-view learning, which will or have started to benefit genomic studies.

Background: While acute kidney injury (AKI) is a common complication in COVID-19, data on post-AKI kidney function recovery and the clinical factors associated with poor kidney function recovery is lacking., Methods: A retrospective multi-centre observational cohort study comprising 12,891 hospitalized patients aged 18 years or older with a diagnosis of SARS-CoV-2 infection confirmed by polymerase chain reaction from 1 January 2020 to 10 September 2020, and with at least one serum creatinine value 1-365 days prior to admission. Mortality and serum creatinine values were obtained up to 10 September 2021., Findings: Advanced age (HR 2.77, 95%CI 2.53-3.04, p < 0.0001), severe COVID-19 (HR 2.91, 95%CI 2.03-4.17, p < 0.0001), severe AKI (KDIGO stage 3: HR 4.22, 95%CI 3.55-5.00, p < 0.0001), and ischemic heart disease (HR 1.26, 95%CI 1.14-1.39, p < 0.0001) were associated with worse mortality outcomes. AKI severity (KDIGO stage 3: HR 0.41, 95%CI 0.37-0.46, p < 0.0001) was associated with worse kidney function recovery, whereas remdesivir use (HR 1.34, 95%CI 1.17-1.54, p < 0.0001) was associated with better kidney function recovery. In a subset of patients without chronic kidney disease, advanced age (HR 1.38, 95%CI 1.20-1.58, p < 0.0001), male sex (HR 1.67, 95%CI 1.45-1.93, p < 0.0001), severe AKI (KDIGO stage 3: HR 11.68, 95%CI 9.80-13.91, p < 0.0001), and hypertension (HR 1.22, 95%CI 1.10-1.36, p = 0.0002) were associated with post-AKI kidney function impairment. Furthermore, patients with COVID-19-associated AKI had significant and persistent elevations of baseline serum creatinine 125% or more at 180 days (RR 1.49, 95%CI 1.32-1.67) and 365 days (RR 1.54, 95%CI 1.21-1.96) compared to COVID-19 patients with no AKI., Interpretation: COVID-19-associated AKI was associated with higher mortality, and severe COVID-19-associated AKI was associated with worse long-term post-AKI kidney function recovery., Funding: Authors are supported by various funders, with full details stated in the acknowledgement section., Competing Interests: Dr Hanauer reported having developed an electronic resource of clinical synonyms, EMERSE, that is licensed by the University of Michigan and receiving a portion of the licensing fees for this resource outside the submitted work. Dr Omenn reported being an early investor and serving on the board of Angion Biomedica Corporation, New York, which has conducted clinical trials of drug candidates for overcoming acute kidney injury following cardiopulmonary surgery or kidney transplantation. The former was terminated early based on unsatisfactory efficacy/adverse effects assessment; the latter had insufficient benefit to warrant proposing a Phase III trial. The company is moving in other directions, to be determined. No further work on kidney is anticipated. Dr Holmes disclosed participation as an NIH/NIDDK T2 Coach R01DK113189. Dr Malovini disclosed being a shareholder of Biomeris s.r.l. Dr Bellazzi reported receiving honoraria from Pfizer, and disclosed being a shareholder of University of Pavia spin-off Biomeris. Dr Klann reports consulting fees from i2b2 tranSMART foundation, for work to enhance open-source data warehouse platform. He reports no direct relationship to this work, except that the data model for analysis in this manuscript was inspired by this platform., (© 2022 Published by Elsevier Ltd.)

The risk profiles of post-acute sequelae of COVID-19 (PASC) have not been well characterized in multi-national settings with appropriate controls. We leveraged electronic health record (EHR) data from 277 international hospitals representing 414,602 patients with COVID-19, 2.3 million control patients without COVID-19 in the inpatient and outpatient settings, and over 221 million diagnosis codes to systematically identify new-onset conditions enriched among patients with COVID-19 during the post-acute period. Compared to inpatient controls, inpatient COVID-19 cases were at significant risk for angina pectoris (RR 1.30, 95% CI 1.09-1.55), heart failure (RR 1.22, 95% CI 1.10-1.35), cognitive dysfunctions (RR 1.18, 95% CI 1.07-1.31), and fatigue (RR 1.18, 95% CI 1.07-1.30). Relative to outpatient controls, outpatient COVID-19 cases were at risk for pulmonary embolism (RR 2.10, 95% CI 1.58-2.76), venous embolism (RR 1.34, 95% CI 1.17-1.54), atrial fibrillation (RR 1.30, 95% CI 1.13-1.50), type 2 diabetes (RR 1.26, 95% CI 1.16-1.36) and vitamin D deficiency (RR 1.19, 95% CI 1.09-1.30). Outpatient COVID-19 cases were also at risk for loss of smell and taste (RR 2.42, 95% CI 1.90-3.06), inflammatory neuropathy (RR 1.66, 95% CI 1.21-2.27), and cognitive dysfunction (RR 1.18, 95% CI 1.04-1.33). The incidence of post-acute cardiovascular and pulmonary conditions decreased across time among inpatient cases while the incidence of cardiovascular, digestive, and metabolic conditions increased among outpatient cases. Our study, based on a federated international network, systematically identified robust conditions associated with PASC compared to control groups, underscoring the multifaceted cardiovascular and neurological phenotype profiles of PASC., (© 2022. The Author(s).)

Objective: To assess changes in international mortality rates and laboratory recovery rates during hospitalisation for patients hospitalised with SARS-CoV-2 between the first wave (1 March to 30 June 2020) and the second wave (1 July 2020 to 31 January 2021) of the COVID-19 pandemic., Design, Setting and Participants: This is a retrospective cohort study of 83 178 hospitalised patients admitted between 7 days before or 14 days after PCR-confirmed SARS-CoV-2 infection within the Consortium for Clinical Characterization of COVID-19 by Electronic Health Record, an international multihealthcare system collaborative of 288 hospitals in the USA and Europe. The laboratory recovery rates and mortality rates over time were compared between the two waves of the pandemic., Primary and Secondary Outcome Measures: The primary outcome was all-cause mortality rate within 28 days after hospitalisation stratified by predicted low, medium and high mortality risk at baseline. The secondary outcome was the average rate of change in laboratory values during the first week of hospitalisation., Results: Baseline Charlson Comorbidity Index and laboratory values at admission were not significantly different between the first and second waves. The improvement in laboratory values over time was faster in the second wave compared with the first. The average C reactive protein rate of change was -4.72 mg/dL vs -4.14 mg/dL per day (p=0.05). The mortality rates within each risk category significantly decreased over time, with the most substantial decrease in the high-risk group (42.3% in March-April 2020 vs 30.8% in November 2020 to January 2021, p<0.001) and a moderate decrease in the intermediate-risk group (21.5% in March-April 2020 vs 14.3% in November 2020 to January 2021, p<0.001)., Conclusions: Admission profiles of patients hospitalised with SARS-CoV-2 infection did not differ greatly between the first and second waves of the pandemic, but there were notable differences in laboratory improvement rates during hospitalisation. Mortality risks among patients with similar risk profiles decreased over the course of the pandemic. The improvement in laboratory values and mortality risk was consistent across multiple countries., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Given the growing number of prediction algorithms developed to predict COVID-19 mortality, we evaluated the transportability of a mortality prediction algorithm using a multi-national network of healthcare systems. We predicted COVID-19 mortality using baseline commonly measured laboratory values and standard demographic and clinical covariates across healthcare systems, countries, and continents. Specifically, we trained a Cox regression model with nine measured laboratory test values, standard demographics at admission, and comorbidity burden pre-admission. These models were compared at site, country, and continent level. Of the 39,969 hospitalized patients with COVID-19 (68.6% male), 5717 (14.3%) died. In the Cox model, age, albumin, AST, creatine, CRP, and white blood cell count are most predictive of mortality. The baseline covariates are more predictive of mortality during the early days of COVID-19 hospitalization. Models trained at healthcare systems with larger cohort size largely retain good transportability performance when porting to different sites. The combination of routine laboratory test values at admission along with basic demographic features can predict mortality in patients hospitalized with COVID-19. Importantly, this potentially deployable model differs from prior work by demonstrating not only consistent performance but also reliable transportability across healthcare systems in the US and Europe, highlighting the generalizability of this model and the overall approach., (© 2022. The Author(s).)

Importance: More than 1 billion adults have hypertension globally, of whom 70% cannot achieve their hypertension control goal with monotherapy alone. Data are lacking on clinical use patterns of dual combination therapies prescribed to patients who escalate from monotherapy., Objective: To investigate the most common dual combinations prescribed for treatment escalation in different countries and how treatment use varies by age, sex, and history of cardiovascular disease., Design, Setting, and Participants: This cohort study used data from 11 electronic health record databases that cover 118 million patients across 8 countries and regions between January 2000 and December 2019. Included participants were adult patients (ages ≥18 years) who newly initiated antihypertensive dual combination therapy after escalating from monotherapy. There were 2 databases included for 3 countries: the Iqvia Longitudinal Patient Database (LPD) Australia and Electronic Practice-based Research Network 2019 linked data set from South Western Sydney Local Health District (ePBRN SWSLHD) from Australia, Ajou University School of Medicine (AUSOM) and Kyung Hee University Hospital (KHMC) databases from South Korea, and Khoo Teck Puat Hospital (KTPH) and National University Hospital (NUH) databases from Singapore. Data were analyzed from June 2020 through August 2021., Exposures: Treatment with dual combinations of the 4 most commonly used antihypertensive drug classes (angiotensin-converting enzyme inhibitor [ACEI] or angiotensin receptor blocker [ARB]; calcium channel blocker [CCB]; β-blocker; and thiazide or thiazide-like diuretic)., Main Outcomes and Measures: The proportion of patients receiving each dual combination regimen, overall and by country and demographic subgroup., Results: Among 970 335 patients with hypertension who newly initiated dual combination therapy included in the final analysis, there were 11 494 patients from Australia (including 9291 patients in Australia LPD and 2203 patients in ePBRN SWSLHD), 6980 patients from South Korea (including 6029 patients in Ajou University and 951 patients in KHMC), 2096 patients from Singapore (including 842 patients in KTPH and 1254 patients in NUH), 7008 patients from China, 8544 patients from Taiwan, 103 994 patients from France, 76 082 patients from Italy, and 754 137 patients from the US. The mean (SD) age ranged from 57.6 (14.8) years in China to 67.7 (15.9) years in the Singapore KTPH database, and the proportion of patients by sex ranged from 24 358 (36.9%) women in Italy to 408 964 (54.3%) women in the US. Among 12 dual combinations of antihypertensive drug classes commonly used, there were significant variations in use across country and patient subgroup. For example starting an ACEI or ARB monotherapy followed by a CCB (ie, ACEI or ARB + CCB) was the most commonly prescribed combination in Australia (698 patients in ePBRN SWSLHD [31.7%] and 3842 patients in Australia LPD [41.4%]) and Singapore (216 patients in KTPH [25.7%] and 439 patients in NUH [35.0%]), while in South Korea, CCB + ACEI or ARB (191 patients in KHMC [20.1%] and 1487 patients in Ajou University [24.7%]), CCB + β-blocker (814 patients in Ajou University [13.5%] and 217 patients in KHMC [22.8%]), and ACEI or ARB + CCB (147 patients in KHMC [15.5%] and 1216 patients in Ajou University [20.2%]) were the 3 most commonly prescribed combinations. The distribution of 12 dual combination therapies were significantly different by age and sex in almost all databases. For example, use of ACEI or ARB + CCB varied from 873 of 3737 patients ages 18 to 64 years (23.4%) to 343 of 2292 patients ages 65 years or older (15.0%) in South Korea's Ajou University database (P for database distribution by age < .001), while use of ACEI or ARB + CCB varied from 2121 of 4718 (44.8%) men to 1721 of 4549 (37.7%) women in Australian LPD (P for drug combination distributions by sex < .001)., Conclusions and Relevance: In this study, large variation in the transition between monotherapy and dual combination therapy for hypertension was observed across countries and by demographic group. These findings suggest that future research may be needed to investigate what dual combinations are associated with best outcomes for which patients.

Background: Acute kidney injury (AKI) develops in 4% of hospitalized patients and is a marker of clinical deterioration and nephrotoxicity. AKI onset is highly variable in hospitals, which makes it difficult to time biomarker assessment in all patients for preemptive care., Objective: The study sought to apply machine learning techniques to electronic health records and predict hospital-acquired AKI by a 48-hour lead time, with the aim to create an AKI surveillance algorithm that is deployable in real time., Methods: The data were sourced from 20,732 case admissions in 16,288 patients over 1 year in our institution. We enhanced the bidirectional recurrent neural network model with a novel time-invariant and time-variant aggregated module to capture important clinical features temporal to AKI in every patient. Time-series features included laboratory parameters that preceded a 48-hour prediction window before AKI onset; the latter's corresponding reference was the final in-hospital serum creatinine performed in case admissions without AKI episodes., Results: The cohort was of mean age 53 (SD 25) years, of whom 29%, 12%, 12%, and 53% had diabetes, ischemic heart disease, cancers, and baseline eGFR <90 mL/min/1.73 m 2 , respectively. There were 911 AKI episodes in 869 patients. We derived and validated an algorithm in the testing dataset with an AUROC of 0.81 (0.78-0.85) for predicting AKI. At a 15% prediction threshold, our model generated 699 AKI alerts with 2 false positives for every true AKI and predicted 26% of AKIs. A lowered 5% prediction threshold improved the recall to 60% but generated 3746 AKI alerts with 6 false positives for every true AKI. Representative interpretation results produced by our model alluded to the top-ranked features that predicted AKI that could be categorized in association with sepsis, acute coronary syndrome, nephrotoxicity, or multiorgan injury, specific to every case at risk., Conclusions: We generated an accurate algorithm from electronic health records through machine learning that predicted AKI by a lead time of at least 48 hours. The prediction threshold could be adjusted during deployment to optimize recall and minimize alert fatigue, while its precision could potentially be augmented by targeted AKI biomarker assessment in the high-risk cohort identified., (©Horng-Ruey Chua, Kaiping Zheng, Anantharaman Vathsala, Kee-Yuan Ngiam, Hui-Kim Yap, Liangjian Lu, Ho-Yee Tiong, Amartya Mukhopadhyay, Graeme MacLaren, Shir-Lynn Lim, K Akalya, Beng-Chin Ooi. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 24.12.2021.)

Background . In critical care, intensivists are required to continuously monitor high-dimensional vital signs and lab measurements to detect and diagnose acute patient conditions, which has always been a challenging task. Recently, deep learning models such as recurrent neural networks (RNNs) have demonstrated their strong potential on predicting such events. However, in real deployment, the patient data are continuously coming and there is no effective adaptation mechanism for RNN to incorporate those new data and become more accurate. Methods . In this study, we propose a novel self-correcting mechanism for RNN to fill in this gap. Our mechanism feeds prediction errors from the predictions of previous timestamps into the prediction of the current timestamp, so that the model can "learn" from previous predictions. We also proposed a regularization method that takes into account not only the model's prediction errors on the labels but also its estimation errors on the input data. Results . We compared the performance of our proposed method with the conventional deep learning models on two real-world clinical datasets for the task of acute kidney injury (AKI) prediction and demonstrated that the proposed model achieved an area under ROC curve at 0.893 on the MIMIC-III dataset and 0.871 on the Philips eICU dataset. Conclusions . The proposed self-correcting RNNs demonstrated effectiveness in AKI prediction and have the potential to be applied to clinical applications., Competing Interests: The authors declare that there is no conflict of interest., (Copyright © 2021 Hao Du et al.)

Objective: Mobile-based interventions have the potential to promote healthy aging among older adults. However, the adoption and use of mobile health applications are often low due to inappropriate designs. The aim of this systematic review is to identify, synthesize, and report interface and persuasive feature design recommendations of mobile health applications for elderly users to facilitate adoption and improve health-related outcomes., Materials and Methods: We searched PubMed, Embase, PsycINFO, CINAHL, and Scopus databases to identify studies that discussed and evaluated elderly-friendly interface and persuasive feature designs of mobile health applications using an elderly cohort., Results: We included 74 studies in our analysis. Our analysis revealed a total of 9 elderly-friendly interface design recommendations: 3 recommendations were targeted at perceptual capabilities of elderly users, 2 at motor coordination problems, and 4 at cognitive and memory deterioration. We also compiled and reported 5 categories of persuasive features: reminders, social features, game elements, personalized interventions, and health education., Discussion: Only 5 studies included design elements that were based on theories. Moreover, the majority of the included studies evaluated the application as a whole without examining end-user perceptions and the effectiveness of each single design feature. Finally, most studies had methodological limitations, and better research designs are needed to quantify the effectiveness of the application designs rigorously., Conclusions: This review synthesizes elderly-friendly interface and persuasive feature design recommendations for mobile health applications from the existing literature and provides recommendations for future research in this area and guidelines for designers., (© The Author(s) 2021. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Neurological complications worsen outcomes in COVID-19. To define the prevalence of neurological conditions among hospitalized patients with a positive SARS-CoV-2 reverse transcription polymerase chain reaction test in geographically diverse multinational populations during early pandemic, we used electronic health records (EHR) from 338 participating hospitals across 6 countries and 3 continents (January-September 2020) for a cross-sectional analysis. We assessed the frequency of International Classification of Disease code of neurological conditions by countries, healthcare systems, time before and after admission for COVID-19 and COVID-19 severity. Among 35,177 hospitalized patients with SARS-CoV-2 infection, there was an increase in the proportion with disorders of consciousness (5.8%, 95% confidence interval [CI] 3.7-7.8%, p FDR  < 0.001) and unspecified disorders of the brain (8.1%, 5.7-10.5%, p FDR  < 0.001) when compared to the pre-admission proportion. During hospitalization, the relative risk of disorders of consciousness (22%, 19-25%), cerebrovascular diseases (24%, 13-35%), nontraumatic intracranial hemorrhage (34%, 20-50%), encephalitis and/or myelitis (37%, 17-60%) and myopathy (72%, 67-77%) were higher for patients with severe COVID-19 when compared to those who never experienced severe COVID-19. Leveraging a multinational network to capture standardized EHR data, we highlighted the increased prevalence of central and peripheral neurological phenotypes in patients hospitalized with COVID-19, particularly among those with severe disease., (© 2021. The Author(s).)

Background: Many countries have experienced 2 predominant waves of COVID-19-related hospitalizations. Comparing the clinical trajectories of patients hospitalized in separate waves of the pandemic enables further understanding of the evolving epidemiology, pathophysiology, and health care dynamics of the COVID-19 pandemic., Objective: In this retrospective cohort study, we analyzed electronic health record (EHR) data from patients with SARS-CoV-2 infections hospitalized in participating health care systems representing 315 hospitals across 6 countries. We compared hospitalization rates, severe COVID-19 risk, and mean laboratory values between patients hospitalized during the first and second waves of the pandemic., Methods: Using a federated approach, each participating health care system extracted patient-level clinical data on their first and second wave cohorts and submitted aggregated data to the central site. Data quality control steps were adopted at the central site to correct for implausible values and harmonize units. Statistical analyses were performed by computing individual health care system effect sizes and synthesizing these using random effect meta-analyses to account for heterogeneity. We focused the laboratory analysis on C-reactive protein (CRP), ferritin, fibrinogen, procalcitonin, D-dimer, and creatinine based on their reported associations with severe COVID-19., Results: Data were available for 79,613 patients, of which 32,467 were hospitalized in the first wave and 47,146 in the second wave. The prevalence of male patients and patients aged 50 to 69 years decreased significantly between the first and second waves. Patients hospitalized in the second wave had a 9.9% reduction in the risk of severe COVID-19 compared to patients hospitalized in the first wave (95% CI 8.5%-11.3%). Demographic subgroup analyses indicated that patients aged 26 to 49 years and 50 to 69 years; male and female patients; and black patients had significantly lower risk for severe disease in the second wave than in the first wave. At admission, the mean values of CRP were significantly lower in the second wave than in the first wave. On the seventh hospital day, the mean values of CRP, ferritin, fibrinogen, and procalcitonin were significantly lower in the second wave than in the first wave. In general, countries exhibited variable changes in laboratory testing rates from the first to the second wave. At admission, there was a significantly higher testing rate for D-dimer in France, Germany, and Spain., Conclusions: Patients hospitalized in the second wave were at significantly lower risk for severe COVID-19. This corresponded to mean laboratory values in the second wave that were more likely to be in typical physiological ranges on the seventh hospital day compared to the first wave. Our federated approach demonstrated the feasibility and power of harmonizing heterogeneous EHR data from multiple international health care systems to rapidly conduct large-scale studies to characterize how COVID-19 clinical trajectories evolve., (©Griffin M Weber, Harrison G Zhang, Sehi L'Yi, Clara-Lea Bonzel, Chuan Hong, Paul Avillach, Alba Gutiérrez-Sacristán, Nathan P Palmer, Amelia Li Min Tan, Xuan Wang, William Yuan, Nils Gehlenborg, Anna Alloni, Danilo F Amendola, Antonio Bellasi, Riccardo Bellazzi, Michele Beraghi, Mauro Bucalo, Luca Chiovato, Kelly Cho, Arianna Dagliati, Hossein Estiri, Robert W Follett, Noelia García Barrio, David A Hanauer, Darren W Henderson, Yuk-Lam Ho, John H Holmes, Meghan R Hutch, Ramakanth Kavuluru, Katie Kirchoff, Jeffrey G Klann, Ashok K Krishnamurthy, Trang T Le, Molei Liu, Ne Hooi Will Loh, Sara Lozano-Zahonero, Yuan Luo, Sarah Maidlow, Adeline Makoudjou, Alberto Malovini, Marcelo Roberto Martins, Bertrand Moal, Michele Morris, Danielle L Mowery, Shawn N Murphy, Antoine Neuraz, Kee Yuan Ngiam, Marina P Okoshi, Gilbert S Omenn, Lav P Patel, Miguel Pedrera Jiménez, Robson A Prudente, Malarkodi Jebathilagam Samayamuthu, Fernando J Sanz Vidorreta, Emily R Schriver, Petra Schubert, Pablo Serrano Balazote, Byorn WL Tan, Suzana E Tanni, Valentina Tibollo, Shyam Visweswaran, Kavishwar B Wagholikar, Zongqi Xia, Daniela Zöller, The Consortium For Clinical Characterization Of COVID-19 By EHR (4CE), Isaac S Kohane, Tianxi Cai, Andrew M South, Gabriel A Brat. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 11.10.2021.)

Background: The standard of care for thyroid cancer management is thyroidectomy and adjuvant radioactive iodine (RAI). There is a paucity of clinical tool that quantifies residual thyroid volume reliably for precise adjuvant RAI dosing. Serum thyroglobulin (TG), tumour marker for thyroid cancer, takes 4 weeks for complete clearance due to its long half-life, and might be undetectable in 12% of structural disease patients. It detects recurrence with a sensitivity of 19-40%, mainly attributed to issue of TG antibody interference with TG immunometric assay. We hypothesise that the quantity of thyroid-specific cell-free RNA (cfRNA) is indicative of amount of thyroid tissues, and that during thyroid surgery, cfRNA levels decrease accordingly., Methods: We identified 11 biologically significant and highly expressed thyroid-specific targets from Human Protein Atlas and literature. To assess for a fall in thyroid-specific cfRNA level, we recruited 16 patients undergoing thyroid surgery or RAI for malignant or benign thyroid disease, and tracked longitudinal trend of cfRNA. To assess the utility of cfRNA in detecting metastatic thyroid cancer, cfRNA of 11 patients at intermediate to high risk of recurrence was measured during surveillance and at time of clinical recurrence., Results: The multiplex assay was capable of amplifying and quantifying multiple thyroid-specific genes in a single reaction. The selected targets were amplified successfully from RNA extracted directly from the thyroid (positive control), indicating that they were highly expressed within thyroid tissue. These cfRNAs were present in plasma, in amounts quantifiable using qRT-PCR. Four cfRNA transcripts (TPO, GFRA2, IVD, TG) fell post-treatment in more than 50% of cohort. The thyroid peroxidase (TPO) cfRNA fell post-therapy in 63% of cohort by 80%, as early as 1 day post-treatment, supporting the potential role as early indicator of remnant thyroid tissue volume. We demonstrated the clinical relevance of circulating TPO cfRNA by tracking temporal changes in setting of peri-treatment, recurrence, and TG Ab positive state., Conclusion: Using a multiplex pre-amplification approach, the TPO cfRNA was a potential biomarker that can track residual thyroid mass. It can be further optimised for quantification of thyroid volume to guide RAI doses and for detection of thyroid cancer recurrence., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Yang, Koh, Kong, Parameswaran, Loke, Ngiam, Tan, Loh, Ng, Goh, Ngeow and Tai.)

Objective: The Consortium for Clinical Characterization of COVID-19 by EHR (4CE) is an international collaboration addressing coronavirus disease 2019 (COVID-19) with federated analyses of electronic health record (EHR) data. We sought to develop and validate a computable phenotype for COVID-19 severity., Materials and Methods: Twelve 4CE sites participated. First, we developed an EHR-based severity phenotype consisting of 6 code classes, and we validated it on patient hospitalization data from the 12 4CE clinical sites against the outcomes of intensive care unit (ICU) admission and/or death. We also piloted an alternative machine learning approach and compared selected predictors of severity with the 4CE phenotype at 1 site., Results: The full 4CE severity phenotype had pooled sensitivity of 0.73 and specificity 0.83 for the combined outcome of ICU admission and/or death. The sensitivity of individual code categories for acuity had high variability-up to 0.65 across sites. At one pilot site, the expert-derived phenotype had mean area under the curve of 0.903 (95% confidence interval, 0.886-0.921), compared with an area under the curve of 0.956 (95% confidence interval, 0.952-0.959) for the machine learning approach. Billing codes were poor proxies of ICU admission, with as low as 49% precision and recall compared with chart review., Discussion: We developed a severity phenotype using 6 code classes that proved resilient to coding variability across international institutions. In contrast, machine learning approaches may overfit hospital-specific orders. Manual chart review revealed discrepancies even in the gold-standard outcomes, possibly owing to heterogeneous pandemic conditions., Conclusions: We developed an EHR-based severity phenotype for COVID-19 in hospitalized patients and validated it at 12 international sites., (© The Author(s) 2021. Published by Oxford University Press on behalf of the American Medical Informatics Association.)

Importance: Additional sources of pediatric epidemiological and clinical data are needed to efficiently study COVID-19 in children and youth and inform infection prevention and clinical treatment of pediatric patients., Objective: To describe international hospitalization trends and key epidemiological and clinical features of children and youth with COVID-19., Design, Setting, and Participants: This retrospective cohort study included pediatric patients hospitalized between February 2 and October 10, 2020. Patient-level electronic health record (EHR) data were collected across 27 hospitals in France, Germany, Spain, Singapore, the UK, and the US. Patients younger than 21 years who tested positive for COVID-19 and were hospitalized at an institution participating in the Consortium for Clinical Characterization of COVID-19 by EHR were included in the study., Main Outcomes and Measures: Patient characteristics, clinical features, and medication use., Results: There were 347 males (52%; 95% CI, 48.5-55.3) and 324 females (48%; 95% CI, 44.4-51.3) in this study's cohort. There was a bimodal age distribution, with the greatest proportion of patients in the 0- to 2-year (199 patients [30%]) and 12- to 17-year (170 patients [25%]) age range. Trends in hospitalizations for 671 children and youth found discrete surges with variable timing across 6 countries. Data from this cohort mirrored national-level pediatric hospitalization trends for most countries with available data, with peaks in hospitalizations during the initial spring surge occurring within 23 days in the national-level and 4CE data. A total of 27 364 laboratory values for 16 laboratory tests were analyzed, with mean values indicating elevations in markers of inflammation (C-reactive protein, 83 mg/L; 95% CI, 53-112 mg/L; ferritin, 417 ng/mL; 95% CI, 228-607 ng/mL; and procalcitonin, 1.45 ng/mL; 95% CI, 0.13-2.77 ng/mL). Abnormalities in coagulation were also evident (D-dimer, 0.78 ug/mL; 95% CI, 0.35-1.21 ug/mL; and fibrinogen, 477 mg/dL; 95% CI, 385-569 mg/dL). Cardiac troponin, when checked (n = 59), was elevated (0.032 ng/mL; 95% CI, 0.000-0.080 ng/mL). Common complications included cardiac arrhythmias (15.0%; 95% CI, 8.1%-21.7%), viral pneumonia (13.3%; 95% CI, 6.5%-20.1%), and respiratory failure (10.5%; 95% CI, 5.8%-15.3%). Few children were treated with COVID-19-directed medications., Conclusions and Relevance: This study of EHRs of children and youth hospitalized for COVID-19 in 6 countries demonstrated variability in hospitalization trends across countries and identified common complications and laboratory abnormalities in children and youth with COVID-19 infection. Large-scale informatics-based approaches to integrate and analyze data across health care systems complement methods of disease surveillance and advance understanding of epidemiological and clinical features associated with COVID-19 in children and youth.

Background: Artificial intelligence (AI) applications are growing at an unprecedented pace in health care, including disease diagnosis, triage or screening, risk analysis, surgical operations, and so forth. Despite a great deal of research in the development and validation of health care AI, only few applications have been actually implemented at the frontlines of clinical practice., Objective: The objective of this study was to systematically review AI applications that have been implemented in real-life clinical practice., Methods: We conducted a literature search in PubMed, Embase, Cochrane Central, and CINAHL to identify relevant articles published between January 2010 and May 2020. We also hand searched premier computer science journals and conferences as well as registered clinical trials. Studies were included if they reported AI applications that had been implemented in real-world clinical settings., Results: We identified 51 relevant studies that reported the implementation and evaluation of AI applications in clinical practice, of which 13 adopted a randomized controlled trial design and eight adopted an experimental design. The AI applications targeted various clinical tasks, such as screening or triage (n=16), disease diagnosis (n=16), risk analysis (n=14), and treatment (n=7). The most commonly addressed diseases and conditions were sepsis (n=6), breast cancer (n=5), diabetic retinopathy (n=4), and polyp and adenoma (n=4). Regarding the evaluation outcomes, we found that 26 studies examined the performance of AI applications in clinical settings, 33 studies examined the effect of AI applications on clinician outcomes, 14 studies examined the effect on patient outcomes, and one study examined the economic impact associated with AI implementation., Conclusions: This review indicates that research on the clinical implementation of AI applications is still at an early stage despite the great potential. More research needs to assess the benefits and challenges associated with clinical AI applications through a more rigorous methodology., (©Jiamin Yin, Kee Yuan Ngiam, Hock Hai Teo. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 22.04.2021.)

Coincident with the tsunami of COVID-19-related publications, there has been a surge of studies using real-world data, including those obtained from the electronic health record (EHR). Unfortunately, several of these high-profile publications were retracted because of concerns regarding the soundness and quality of the studies and the EHR data they purported to analyze. These retractions highlight that although a small community of EHR informatics experts can readily identify strengths and flaws in EHR-derived studies, many medical editorial teams and otherwise sophisticated medical readers lack the framework to fully critically appraise these studies. In addition, conventional statistical analyses cannot overcome the need for an understanding of the opportunities and limitations of EHR-derived studies. We distill here from the broader informatics literature six key considerations that are crucial for appraising studies utilizing EHR data: data completeness, data collection and handling (eg, transformation), data type (ie, codified, textual), robustness of methods against EHR variability (within and across institutions, countries, and time), transparency of data and analytic code, and the multidisciplinary approach. These considerations will inform researchers, clinicians, and other stakeholders as to the recommended best practices in reviewing manuscripts, grants, and other outputs from EHR-data derived studies, and thereby promote and foster rigor, quality, and reliability of this rapidly growing field., (©Isaac S Kohane, Bruce J Aronow, Paul Avillach, Brett K Beaulieu-Jones, Riccardo Bellazzi, Robert L Bradford, Gabriel A Brat, Mario Cannataro, James J Cimino, Noelia García-Barrio, Nils Gehlenborg, Marzyeh Ghassemi, Alba Gutiérrez-Sacristán, David A Hanauer, John H Holmes, Chuan Hong, Jeffrey G Klann, Ne Hooi Will Loh, Yuan Luo, Kenneth D Mandl, Mohamad Daniar, Jason H Moore, Shawn N Murphy, Antoine Neuraz, Kee Yuan Ngiam, Gilbert S Omenn, Nathan Palmer, Lav P Patel, Miguel Pedrera-Jiménez, Piotr Sliz, Andrew M South, Amelia Li Min Tan, Deanne M Taylor, Bradley W Taylor, Carlo Torti, Andrew K Vallejos, Kavishwar B Wagholikar, The Consortium For Clinical Characterization Of COVID-19 By EHR (4CE), Griffin M Weber, Tianxi Cai. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 02.03.2021.)

The COVID-19 pandemic has taken the world by storm and overwhelmed healthcare institutions even in developed countries. In response, clinical staff and resources have been redeployed to the areas of greatest need, that is, intensive care units and emergency rooms (ER), to reinforce front-line manpower. We introduce the concept of close air support (CAS) to augment ER operations in an efficient, safe and scalable manner. Teams of five comprising two on-site junior ER physicians would be paired with two CAS doctors, who would be off-site but be in constant communication via teleconferencing to render real-time administrative support. They would be supervised by an ER attending. This reduces direct viral exposure to doctors, conserves precious personal protective equipment and allows ER physicians to focus on patient care. Medical students can also be involved in a safe and supervised manner. After 1 month, the average time to patient disposition was halved. General feedback was also positive. CAS improves efficiency and is safe, scalable and sustainable. It has also empowered a previously untapped group of junior clinicians to support front-line medical operations, while simultaneously protecting them from viral exposure. Institutions can consider adopting our novel approach, with modifications made according to their local context., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2020. No commercial re-use. See rights and permissions. Published by BMJ.)

We leveraged the largely untapped resource of electronic health record data to address critical clinical and epidemiological questions about Coronavirus Disease 2019 (COVID-19). To do this, we formed an international consortium (4CE) of 96 hospitals across five countries (www.covidclinical.net). Contributors utilized the Informatics for Integrating Biology and the Bedside (i2b2) or Observational Medical Outcomes Partnership (OMOP) platforms to map to a common data model. The group focused on temporal changes in key laboratory test values. Harmonized data were analyzed locally and converted to a shared aggregate form for rapid analysis and visualization of regional differences and global commonalities. Data covered 27,584 COVID-19 cases with 187,802 laboratory tests. Case counts and laboratory trajectories were concordant with existing literature. Laboratory tests at the time of diagnosis showed hospital-level differences equivalent to country-level variation across the consortium partners. Despite the limitations of decentralized data generation, we established a framework to capture the trajectory of COVID-19 disease in patients and their response to interventions., Competing Interests: Competing interestsR.B. and A.M. are shareholders of Biomeris s.r.l., (© The Author(s) 2020.)

Background: Decision support systems based on reinforcement learning (RL) have been implemented to facilitate the delivery of personalized care. This paper aimed to provide a comprehensive review of RL applications in the critical care setting., Objective: This review aimed to survey the literature on RL applications for clinical decision support in critical care and to provide insight into the challenges of applying various RL models., Methods: We performed an extensive search of the following databases: PubMed, Google Scholar, Institute of Electrical and Electronics Engineers (IEEE), ScienceDirect, Web of Science, Medical Literature Analysis and Retrieval System Online (MEDLINE), and Excerpta Medica Database (EMBASE). Studies published over the past 10 years (2010-2019) that have applied RL for critical care were included., Results: We included 21 papers and found that RL has been used to optimize the choice of medications, drug dosing, and timing of interventions and to target personalized laboratory values. We further compared and contrasted the design of the RL models and the evaluation metrics for each application., Conclusions: RL has great potential for enhancing decision making in critical care. Challenges regarding RL system design, evaluation metrics, and model choice exist. More importantly, further work is required to validate RL in authentic clinical environments., (©Siqi Liu, Kay Choong See, Kee Yuan Ngiam, Leo Anthony Celi, Xingzhi Sun, Mengling Feng. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 20.07.2020.)

Background: The change in two measurements of a continuous outcome can be modelled directly with a linear regression model, or indirectly with a random effects model (REM) of the individual measurements. These methods are susceptible to model misspecifications, which are commonly addressed by applying monotonic transformations (e.g., Box-Cox transformation) to the outcomes. However, transforming the outcomes complicates the data analysis, especially when variable selection is involved. We propose a robust alternative through a novel application of the conditional probit (cprobit) model., Methods: The cprobit model analyzes the ordered outcomes within each subject, making the estimate invariant to monotonic transformation on the outcome. By scaling the estimate from the cprobit model, we obtain the exposure effect on the change in the observed or Box-Cox transformed outcome, pending the adequacy of the normality assumption on the raw or transformed scale., Results: Using simulated data, we demonstrated a similar good performance of the cprobit model and REM with and without transformation, except for some bias from both methods when the Box-Cox transformation was applied to scenarios with small sample size and strong effects. Only the cprobit model was robust to skewed subject-specific intercept terms when a Box-Cox transformation was used. Using two real datasets from the breast cancer and inpatient glycemic variability studies which utilize electronic medical records, we illustrated the application of our proposed robust approach as a seamless three-step workflow that facilitates the use of Box-Cox transformation to address non-normality with a common underlying model., Conclusions: The cprobit model provides a seamless and robust inference on the change in continuous outcomes, and its three-step workflow is implemented in an R package for easy accessibility.

Artificial intelligence (AI) is set to transform healthcare. Key ethical issues to emerge with this transformation encompass the accountability and transparency of the decisions made by AI-based systems, the potential for group harms arising from algorithmic bias and the professional roles and integrity of clinicians. These concerns must be balanced against the imperatives of generating public benefit with more efficient healthcare systems from the vastly higher and accurate computational power of AI. In weighing up these issues, this paper applies the deliberative balancing approach of the Ethics Framework for Big Data in Health and Research (Xafis et al. 2019). The analysis applies relevant values identified from the framework to demonstrate how decision-makers can draw on them to develop and implement AI-assisted support systems into healthcare and clinical practice ethically and responsibly. Please refer to Xafis et al. (2019) in this special issue of the Asian Bioethics Review for more information on how this framework is to be used, including a full explanation of the key values involved and the balancing approach used in the case study at the end of this paper., (© The Author(s) 2019.)

Background: High-intensity focused ultrasound (HIFU) is a recent noninvasive technique of treating thyroid nodules. Our study aims to investigate the efficacy and safety of HIFU in treating benign thyroid nodules., Methods: This is a retrospective analysis of consecutive patients who underwent HIFU of benign thyroid nodules at our institution from July 2017-2018. All procedures were performed by a single surgeon. Patients were evaluated immediately post-procedure, and at subsequent intervals of 1 week, 1 month, 3 months, and 6 months. The primary endpoint was thyroid nodule volume reduction at 6 months posttreatment. Secondary endpoints were post-procedure local complications., Results: Ten patients with 13 thyroid nodules were included. The median follow-up period was 426 days (range 238-573). Mean maximum diameter reduced from 2.6 cm (±0.8) pretreatment to 1.4 cm (±0.7, P < 0.05) 6 months posttreatment. Mean nodule volume reduced from 5.2 cm 3 (±4.2) pretreatment to 1.5 cm 3 (±1.3, P = 0.01) 6 months posttreatment. Mean volume reduction ratio (VRR) at 6 months posttreatment was 63.2% (±22.5, P < 0.05), with volume reduction of ≥50% in 10 of 13 (76.9%) nodules. Two nodules (15.4%) showed size increases from 4 months posttreatment. No patients experienced local skin burns or hematomas. Mean pain scores were 1.5 (±1.2) immediate post-procedure, 0.8 (±1.5) at 1 week, and 0.6 (±1.2) at 1 month post-procedure, respectively, with no reports of pain beyond 1 month. Only two (20.0%) patients had early, temporary posttreatment voice hoarseness., Conclusion: Our study shows HIFU ablation to be efficacious and safe-with significant thyroid nodule volume reductions, and no significant or prolonged local complications.

Background: Thyroxine replacement following a hemithyroidectomy is not commonly discussed during consent for the procedure as the risk of hypothyroidism is perceived to be low., Methods: Retrospective review of 901 patients who underwent hemithyroidectomy at a tertiary referral institution during the period January 2000 to December 2015. The main outcome studied was the overall incidence of hypothyroidism and the associated risk factors., Results: Hypothyroidism developed in 123 (13%) patients and 94 patients (10%) required hormone supplementation over a mean follow up of 21 months (range 1-168 months). Preoperative TSH of more than 2.5 was seen in 38 of 123 (31%) of patients. Presence of diffuse thyroiditis was the only independent risk factor on multivariate analysis (P = 0.002) found to be associated with the development of hypothyroidism., Conclusion: After thyroid lobectomy, approximately one in 10 patients requiring thyroid hormone treatment for hypothyroidism. Presence of diffuse thyroiditis is a significant risk factor for hypothyroidism., (© 2018 Wiley Periodicals, Inc.)

Analysis of big data by machine learning offers considerable advantages for assimilation and evaluation of large amounts of complex health-care data. However, to effectively use machine learning tools in health care, several limitations must be addressed and key issues considered, such as its clinical implementation and ethics in health-care delivery. Advantages of machine learning include flexibility and scalability compared with traditional biostatistical methods, which makes it deployable for many tasks, such as risk stratification, diagnosis and classification, and survival predictions. Another advantage of machine learning algorithms is the ability to analyse diverse data types (eg, demographic data, laboratory findings, imaging data, and doctors' free-text notes) and incorporate them into predictions for disease risk, diagnosis, prognosis, and appropriate treatments. Despite these advantages, the application of machine learning in health-care delivery also presents unique challenges that require data pre-processing, model training, and refinement of the system with respect to the actual clinical problem. Also crucial are ethical considerations, which include medico-legal implications, doctors' understanding of machine learning tools, and data privacy and security. In this Review, we discuss some of the benefits and challenges of big data and machine learning in health care., (Copyright © 2019 Elsevier Ltd. All rights reserved.)

Objective: BRAF mutation is the commonest mutation seen in papillary thyroid cancer (PTC), but its prevalence and clinical significance vary across countries. We aim to evaluate the prevalence and clinico-pathological correlation of BRAF mutation in PTC patients at our centre., Study Design: Retrospective cohort study of 75 consecutive archival thyroid specimens, whereby BRAF mutation was detected using a polymerase chain reaction (PCR) technique and correlated with clinical and pathological features and outcomes., Setting: Tertiary university hospital in Singapore., Participants: A total of 75 consecutive histologically proven archival thyroid specimens from patients who underwent thyroidectomy for PTC were accrued for this study., Main Outcome Measures: Main outcome is to determine the prevalence of the BRAF mutation in our South-East Asian population. Secondary aim is to correlate the mutational status with adverse pathological features like histological variants, multi-focality, lymphovascular invasion and extra-thyroidal extension, clinical features like demographics, TNM stage, recurrence and survival, as well as treatment details like type of surgery performed and radioiodine doses., Results: BRAF mutation was detected in 56% (42/75) of PTC. All but one BRAF-mutated PTC had the BRAFV600E mutation. BRAF-mutated tumours were associated with an advanced T-stage (P = 0.049) and were more likely to have a central neck dissection (P = 0.036). There was no significant correlation between BRAF mutation status and clinical outcomes., Conclusion: The prevalence of BRAF mutation is 56%. BRAF mutation-positive tumours were associated with locally advanced disease, but not poorer survival., (© 2018 John Wiley & Sons Ltd.)