Web-based artificial intelligence to predict cognitive impairment following stroke: A multicenter study.

BACKGROUND AND PURPOSE: Post-stroke cognitive impairment (PSCI) is highly prevalent in modern society. However, there is limited study implying an accurate and explainable machine learning model to predict PSCI. The aim of this study is to develop and validate a web-based artificial intelligence (AI) tool for predicting PSCI. METHODS: The retrospective cohort study design was conducted to develop and validate a web-based prediction model. Adults who experienced a stroke between January 1, 2004, and September 30, 2017, were enrolled, and patients with PSCI were followed up from the stroke index date until their last follow-up. The model's performance metrics, including accuracy, area under the curve (AUC), recall, precision, and F1 score, were compared. RESULTS: A total of 3209 stroke patients were included in the study. The model demonstrated an accuracy of 0.8793, AUC of 0.9200, recall of 0.6332, precision of 0.9664, and F1 score of 0.7651. In the external validation phase, the accuracy improved to 0.9039, AUC to 0.9094, recall to 0.7457, precision to 0.9168, and F1 score to 0.8224. The final model can be accessed at https://psci-calculator.my.id/. CONCLUSION: Our results are able to produce a user-friendly interface that is useful for health practitioners to perform early prediction on PSCI. These findings also suggest that the provided AI model is reliable and can serve as a roadmap for future studies using AI models in a clinical setting.

Facial Mimicry and Doctor-Patient Satisfaction: The Feasibility of Artificial Empathy in a Clinical Video Data.

Good nonverbal communication between doctor and patient is essential for achieving a successful and therapeutic doctor-patient relationship. Increasing evidence has shown that nonverbal communication mimicry, particularly facial mimicry, where one mirrors another's facial expressions, is linked to empathy and emotion recognition. Empathy is also the key driver of patient satisfaction. This study explores how facial expressions and facial mimicry influence doctor-patient satisfaction during a clinical encounter. We used a facial emotion recognition-based artificial empathy model to analyze 315 recorded clinical video data of doctors and patients in a dermatology outpatient clinic. The results show a significant negative correlation between patients' emotions of sadness and neutral and doctor satisfaction, but no correlation between the duration of doctors mimicking patient emotions and patient satisfaction. These findings provide valuable insights into the future design of systems that can further enhance clinician awareness to maintain communication skills in the search for better doctor-patient satisfaction.

Machine Learning Approaches for Predicting Psoriatic Arthritis Risk Using Electronic Medical Records: Population-Based Study.

BACKGROUND: Psoriasis (PsO) is a chronic, systemic, immune-mediated disease with multiorgan involvement. Psoriatic arthritis (PsA) is an inflammatory arthritis that is present in 6%-42% of patients with PsO. Approximately 15% of patients with PsO have undiagnosed PsA. Predicting patients with a risk of PsA is crucial for providing them with early examination and treatment that can prevent irreversible disease progression and function loss. OBJECTIVE: The aim of this study was to develop and validate a prediction model for PsA based on chronological large-scale and multidimensional electronic medical records using a machine learning algorithm. METHODS: This case-control study used Taiwan's National Health Insurance Research Database from January 1, 1999, to December 31, 2013. The original data set was split into training and holdout data sets in an 80:20 ratio. A convolutional neural network was used to develop a prediction model. This model used 2.5-year diagnostic and medical records (inpatient and outpatient) with temporal-sequential information to predict the risk of PsA for a given patient within the next 6 months. The model was developed and cross-validated using the training data and was tested using the holdout data. An occlusion sensitivity analysis was performed to identify the important features of the model. RESULTS: The prediction model included a total of 443 patients with PsA with earlier diagnosis of PsO and 1772 patients with PsO without PsA for the control group. The 6-month PsA risk prediction model that uses sequential diagnostic and drug prescription information as a temporal phenomic map yielded an area under the receiver operating characteristic curve of 0.70 (95% CI 0.559-0.833), a mean sensitivity of 0.80 (SD 0.11), a mean specificity of 0.60 (SD 0.04), and a mean negative predictive value of 0.93 (SD 0.04). CONCLUSIONS: The findings of this study suggest that the risk prediction model can identify patients with PsO at a high risk of PsA. This model may help health care professionals to prioritize treatment for target high-risk populations and prevent irreversible disease progression and functional loss.

Post-Stroke Insomnia Increased the Risk of Cognitive Impairments: A Hospital-Based Retrospective Cohort Study.

OBJECTIVES/BACKGROUND: Insomnia is a common sleep complaint among patients who had a stroke and has been recognized as an independent risk factor for cognitive impairment. However, the relationship between poststroke insomnia and cognitive impairment over time is under-researched. Therefore, we examined the association between poststroke insomnia and the risk of cognitive impairment. PARTICIPANTS: Stroke participants who had a stroke and were 20 years and older. METHODS: This multicenter hospital-based retrospective cohort study with a 13-year follow-up period (2004-2017). The diagnosis of stroke, insomnia, and cognitive impairment was based on the International Classification of Diseases. The study participants who experienced a stroke were divided into two cohorts: those who also had insomnia and those who did not have insomnia. A Cox proportional-hazards regression model was used. RESULTS: A total of 1,775 patients with a mean age of 67.6 years were included. Of these patients, 146 and 75 patients were diagnosed with insomnia and cognitive impairment during the follow-up period, respectively. The cumulative incidence of cognitive impairment in the stroke with insomnia cohort was significantly lower than that in the stroke without insomnia cohort (log-rank test, P < .001). The adjusted hazard ratio and 95% confidence interval (CI) of the stroke with insomnia cohort indicated a higher risk of cognitive impairment compared with the stroke without insomnia cohort (adjusted hazard ratio: 2.38; 95% CI: 1.41-4.03). CONCLUSIONS: Patients who had a stroke and were diagnosed with insomnia exhibited a substantial increased risk of cognitive impairment over time.

SlimMe, a Chatbot With Artificial Empathy for Personal Weight Management: System Design and Finding.

As the obesity rate continues to increase persistently, there is an urgent need to develop an effective weight loss management strategy. Nowadays, the development of artificial intelligence (AI) and cognitive technologies coupled with the rapid spread of messaging platforms and mobile technology with easier access to internet technology offers professional dietitians an opportunity to provide extensive monitoring support to their clients through a chatbot with artificial empathy. This study aimed to design a chatbot with artificial empathic motivational support for weight loss called "SlimMe" and investigate how people react to a diet bot. The SlimMe infrastructure was built using Dialogflow as the natural language processing (NLP) platform and LINE mobile messenger as the messaging platform. We proposed a text-based emotion analysis to simulate artificial empathy responses to recognize the user's emotion. A preliminary evaluation was performed to investigate the early-stage user experience after a 7-day simulation trial. The result revealed that having an artificially empathic diet bot for weight loss management is a fun and exciting experience. The use of emoticons, stickers, and GIF images makes the chatbot response more interactive. Moreover, the motivational support and persuasive messaging features enable the bot to express more empathic and engaging responses to the user. In total, there were 1,007 bot responses from 892 user input messages. Of these, 67.38% (601/1,007) of the chatbot-generated responses were accurate to a relevant user request, 21.19% (189/1,007) inaccurate responses to a relevant request, and 10.31% (92/1,007) accurate responses to an irrelevant request. Only 1.12% (10/1,007) of the chatbot does not answer. We present the design of an artificially empathic diet bot as a friendly assistant to help users estimate their calorie intake and calories burned in a more interactive and engaging way. To our knowledge, this is the first chatbot designed with artificial empathy features, and it looks very promising in promoting long-term weight management. More user interactions and further data training and validation enhancement will improve the bot's in-built knowledge base and emotional intelligence base.

A Deep Learning Model to Predict Knee Osteoarthritis Based on Nonimage Longitudinal Medical Record.

PURPOSE: To develop deep learning model (Deep-KOA) that can predict the risk of knee osteoarthritis (KOA) within the next year by using the previous three years nonimage-based electronic medical record (EMR) data. PATIENTS AND METHODS: We randomly selected information of two million patients from the Taiwan National Health Insurance Research Database (NHIRD) from January 1, 1999 to December 31, 2013. During the study period, 132,594 patients were diagnosed with KOA, while 1,068,464 patients without KOA were chosen randomly as control. We constructed a feature matrix by using the three-year history of sequential diagnoses, drug prescriptions, age, and sex. Deep learning methods of convolutional neural network (CNN) and artificial neural network (ANN) were used together to develop a risk prediction model. We used the area under the receiver operating characteristic (AUROC), sensitivity, specificity, and precision to evaluate the performance of Deep-KOA. Then, we explored the important features using stepwise feature selection. RESULTS: This study included 132,594 KOA patients, 83,111 females (62.68%), 49,483 males (37.32%), mean age 64.2 years, and 1,068,464 non-KOA patients, 545,902 females (51.09%), 522,562 males (48.91%), mean age 51.00 years. The Deep-KOA achieved an overall AUROC, sensitivity, specificity, and precision of 0.97, 0.89, 0.93, and 0.80 respectively. The discriminative analysis of Deep-KOA showed important features from several diseases such as disorders of the eye and adnexa, acute respiratory infection, other metabolic and immunity disorders, and diseases of the musculoskeletal and connective tissue. Age and sex were not found as the most discriminative features, with AUROC of 0.9593 (-0.76% loss) and 0.9644 (-0.25% loss) respectively. Whereas medications including antacid, cough suppressant, and expectorants were identified as discriminative features. CONCLUSION: Deep-KOA was developed to predict the risk of KOA within one year earlier, which may provide clues for clinical decision support systems to target patients with high risk of KOA to get precision prevention program.

Machine Learning Approach to Reduce Alert Fatigue Using a Disease Medication-Related Clinical Decision Support System: Model Development and Validation.

BACKGROUND: Computerized physician order entry (CPOE) systems are incorporated into clinical decision support systems (CDSSs) to reduce medication errors and improve patient safety. Automatic alerts generated from CDSSs can directly assist physicians in making useful clinical decisions and can help shape prescribing behavior. Multiple studies reported that approximately 90%-96% of alerts are overridden by physicians, which raises questions about the effectiveness of CDSSs. There is intense interest in developing sophisticated methods to combat alert fatigue, but there is no consensus on the optimal approaches so far. OBJECTIVE: Our objective was to develop machine learning prediction models to predict physicians' responses in order to reduce alert fatigue from disease medication-related CDSSs. METHODS: We collected data from a disease medication-related CDSS from a university teaching hospital in Taiwan. We considered prescriptions that triggered alerts in the CDSS between August 2018 and May 2019. Machine learning models, such as artificial neural network (ANN), random forest (RF), naïve Bayes (NB), gradient boosting (GB), and support vector machine (SVM), were used to develop prediction models. The data were randomly split into training (80%) and testing (20%) datasets. RESULTS: A total of 6453 prescriptions were used in our model. The ANN machine learning prediction model demonstrated excellent discrimination (area under the receiver operating characteristic curve [AUROC] 0.94; accuracy 0.85), whereas the RF, NB, GB, and SVM models had AUROCs of 0.93, 0.91, 0.91, and 0.80, respectively. The sensitivity and specificity of the ANN model were 0.87 and 0.83, respectively. CONCLUSIONS: In this study, ANN showed substantially better performance in predicting individual physician responses to an alert from a disease medication-related CDSS, as compared to the other models. To our knowledge, this is the first study to use machine learning models to predict physician responses to alerts; furthermore, it can help to develop sophisticated CDSSs in real-world clinical settings.