Geriatric depression and anxiety screening via deep learning using activity tracking and sleep data.

BACKGROUND: Geriatric depression and anxiety have been identified as mood disorders commonly associated with the onset of dementia. Currently, the diagnosis of geriatric depression and anxiety relies on self-reported assessments for primary screening purposes, which is uncomfortable for older adults and can be prone to misreporting. When a more precise diagnosis is needed, additional methods such as in-depth interviews or functional magnetic resonance imaging are used. However, these methods can not only be time-consuming and costly but also require systematic and cost-effective approaches. OBJECTIVE: The main objective of this study was to investigate the feasibility of training an end-to-end deep learning (DL) model by directly inputting time-series activity tracking and sleep data obtained from consumer-grade wrist-worn activity trackers to identify comorbid depression and anxiety. METHODS: To enhance accuracy, the input of the DL model consisted of step counts and sleep stages as time series data, along with minimal depression and anxiety assessment scores as non-time-series data. The basic structure of the DL model was designed to process mixed-input data and perform multi-label-based classification for depression and anxiety. Various DL models, including the convolutional neural network (CNN) and long short-term memory (LSTM), were applied to process the time-series data, and model selection was conducted by comparing the performances of the hyperparameters. RESULTS: This study achieved significant results in the multi-label classification of depression and anxiety, with a Hamming loss score of 0.0946 in the Residual Network (ResNet), by applying a mixed-input DL model based on activity tracking data. The comparison of hyper-parameter performance and the development of various DL models, such as CNN, LSTM, and ResNet contributed to the optimization of time series data processing and achievement of meaningful results. CONCLUSIONS: This study can be considered as the first to develop a mixed-input DL model based on activity tracking data for the multi-label identification of late-life depression and anxiety. The findings of the study demonstrate the feasibility and potential of using consumer-grade wrist-worn activity trackers in conjunction with DL models to improve the identification of comorbid mental health conditions in older adults. The study also established a multi-label classification framework for identifying the complex symptoms of depression and anxiety.

A Standardized Education Program on Deceased Organ and Tissue Donation for Premedical and Medial Students in Korea.

Background: As the imbalance in organ demand and supply is getting worse, <1000 patients waiting for organ transplants die each year in South Korea. To enhance positive attitudes to deceased organ-tissue donation through systematic education, we developed an educational program with delivery pathways for premedical and medical students. Methods: Online and offline self-learning educational materials on deceased organ-tissue donation were generated and posted on the Vitallink Academy YouTube site. Thirty-two pre- and 15 posteducation questionnaires were developed using a web-based survey platform, and conducted before and immediately after the education process. The education proceeded in 3 steps: (1) group study sessions on selected topics, (2) poster submissions by each group and the selection of excellent poster by the organizing committee, and (3) excellent poster presentation and questions and answers. Results: A total of 141 students in the first year of premedical classes at the Seoul National University College of Medicine participated in this program. Only 24.2% of responders agreed that anyone who was diagnosed with brain death should donate. The proportion of students with positive attitudes toward organ-tissue donation increased from 74.7% to 97.7% (P < 0.001) with our education. Likewise, interest in deceased organ-tissue donation-related issues increased from 33.3% to 84.9% (P < 0.001). The expressed willingness for organ-tissue donation also increased from 76.8% to 96.5% (P < 0.001). The proportion of accepting brain death as the determination of death increased from 61.6% to 89.5% (P < 0.001). Moreover, 81.4% changed their approach and planned to register with an organ donor card. Conclusions: In this study, significant improvements were observed in knowledge, awareness, and attitude toward organ-tissue donation with our newly developed co-participatory education program for premedical students. Hence, target-specific education can be regarded as a valuable approach to enhancing public awareness of deceased organ-tissue donation.

Integrative modeling of tumor genomes and epigenomes for enhanced cancer diagnosis by cell-free DNA.

Multi-cancer early detection remains a key challenge in cell-free DNA (cfDNA)-based liquid biopsy. Here, we perform cfDNA whole-genome sequencing to generate two test datasets covering 2125 patient samples of 9 cancer types and 1241 normal control samples, and also a reference dataset for background variant filtering based on 20,529 low-depth healthy samples. An external cfDNA dataset consisting of 208 cancer and 214 normal control samples is used for additional evaluation. Accuracy for cancer detection and tissue-of-origin localization is achieved using our algorithm, which incorporates cancer type-specific profiles of mutation distribution and chromatin organization in tumor tissues as model references. Our integrative model detects early-stage cancers, including those of pancreatic origin, with high sensitivity that is comparable to that of late-stage detection. Model interpretation reveals the contribution of cancer type-specific genomic and epigenomic features. Our methodologies may lay the groundwork for accurate cfDNA-based cancer diagnosis, especially at early stages.

Development and performance evaluation of an artificial intelligence algorithm using cell-free DNA fragment distance for non-invasive prenatal testing (aiD-NIPT).

With advances in next-generation sequencing technology, non-invasive prenatal testing (NIPT) has been widely implemented to detect fetal aneuploidies, including trisomy 21, 18, and 13 (T21, T18, and T13). Most NIPT methods use cell-free DNA (cfDNA) fragment count (FC) in maternal blood. In this study, we developed a novel NIPT method using cfDNA fragment distance (FD) and convolutional neural network-based artificial intelligence algorithm (aiD-NIPT). Four types of aiD-NIPT algorithm (mean, median, interquartile range, and its ensemble) were developed using 2,215 samples. In an analysis of 17,678 clinical samples, all algorithms showed >99.40% accuracy for T21/T18/T13, and the ensemble algorithm showed the best performance (sensitivity: 99.07%, positive predictive value (PPV): 88.43%); the FC-based conventional Z-score and normalized chromosomal value showed 98.15% sensitivity, with 40.77% and 36.81% PPV, respectively. In conclusion, FD-based aiD-NIPT was successfully developed, and it showed better performance than FC-based NIPT methods.

The Impact of Daily Hours of Sound Therapy on Tinnitus Relief for People With Chronic Tinnitus: A Randomized Controlled Study.

PURPOSE: Tinnitus is defined as the perception of sounds in the absence of extrinsic sound stimuli. Sound therapy is an option for tinnitus rehabilitation, which aims to mitigate the functional and emotional effects of tinnitus. Several studies have reported that a longer duration of sound therapy may result in a greater tinnitus relief effect. However, the relationship between the daily hours of sound therapy and the tinnitus relief effect remains unclear. Thus, we aimed to evaluate the tinnitus relief effect according to the time spent in daily performing sound therapy. METHOD: Fifty-eight chronic tinnitus participants were randomly assigned to three groups according to their daily hours of sound therapy (1-hr, 3-hr, and 5-hr groups), and mixing point-based sound therapy, which adjusts the volume of the external stimulus to blend with the participant's tinnitus sound, was administered for 3 months. The effectiveness of sound therapy was determined based on changes in the tinnitus loudness level, visual analog scale for loudness score, and the Korean version of the Tinnitus Primary Function Questionnaire (K-TPFQ) score between the baseline and 3-month time points. RESULTS: Longer daily sound therapy hours were associated to a greater effect of sound therapy in the tinnitus loudness level and the total K-TPFQ scores, including the emotion and hearing subcategories. Particularly, the 3-hr and 5-hr groups showed a greater tinnitus relief than the 1-hr group. CONCLUSION: An increase in the daily sound therapy hours may increase the possibility of improving tinnitus relief effects.

Pharmacokinetics and Genetic Factors of Atorvastatin in Healthy Korean Subjects.

Background: Statins are the most popular agents for the primary and secondary prevention of cardiovascular disease; however, the pharmacokinetic parameters and associated genetic factors in the Korean population have not been fully elucidated. This study explored the pharmacokinetic properties of atorvastatin and the association between genetic variations and atorvastatin pharmacokinetics in healthy Korean subjects. Methods: Atorvastatin (80 mg) was administered to 35 healthy Korean volunteers. Plasma levels of atorvastatin and its metabolites were measured sequentially using liquid chromatography-tandem mass spectrometry from 0 to 24 h after atorvastatin administration. Customized next-generation sequencing analysis was performed covering all coding exons of 15 genes, as well as 46 single-nucleotide variants in 29 genes related to statin pharmacokinetics. Results: The mean area under the concentration-time (AUC) and Cmax (maximum peak concentration) were 269.0 ng/ml∙h and 84.3 ng/ml, respectively, which were approximately two times higher than those reported in Caucasians. Genetic analysis revealed that eight genetic variants in ABCB1, ABCG2, APOA5, CETP, and CYP7A1 contributed to the AUC of atorvastatin. The atorvastatin AUC0-24 h prediction model was developed based on age and eight genetic variants using multivariate linear regression (adjusted R 2 = 0.878, p < 0.0001). Conclusion: This study shows that the pharmacokinetic properties of atorvastatin in Koreans are different from those in Caucasians and that atorvastatin AUC0-24 h could be predicted based on age and eight genetic variants of ABCB1, ABCG2, APOA5, CETP, and CYP7A1.

Genomic Instability of Circulating Tumor DNA as a Prognostic Marker for Pancreatic Cancer Survival: A Prospective Cohort Study.

Genomic instability of circulating tumor DNA (ctDNA) as a prognostic biomarker has not been evaluated in pancreatic cancer. We investigated the role of the genomic instability index of ctDNA in pancreatic ductal adenocarcinoma (PDAC). We prospectively enrolled 315 patients newly diagnosed with resectable (n = 110), locally advanced (n = 78), and metastatic (n = 127) PDAC from March 2015 through January 2020. Low-depth whole-genome cell-free DNA sequencing identified genome-wide copy number alterations using instability score (I-score) to reflect genome-wide instability. Plasma cell-free and matched tumor tissue DNA from 15 patients with resectable pancreatic cancer was sequenced to assess the concordance of chromosomal copy number alteration profiles. Associations of I-score with clinical factors or survival were assessed. Seventy-six patients had high genomic instability with I-score > 7.3 in pre-treatment ctDNA; proportions of high I-score were 5.5%, 5.1%, and 52% in resectable, locally advanced, and metastatic stages, respectively. Correlation coefficients between Z-scores of plasma and tissue DNA at segment resolution were high (r2 = 0.82). Univariable analysis showed the association of I-score with progression-free survival in each stage. Multivariable analyses demonstrated that clinical stage-adjusted I-scores were significant factors for progression-free and overall survival. In these patients, ctDNA genomic I-scores provided prognostic information relevant to progression-free survival in each clinical stage.

A Real-Time Autonomous Dashboard for the Emergency Department: 5-Year Case Study.

BACKGROUND: The task of monitoring and managing the entire emergency department (ED) is becoming more important due to increasing pressure on the ED. Recently, dashboards have received the spotlight as health information technology to support these tasks. OBJECTIVE: This study aimed to describe the development of a real-time autonomous dashboard for the ED and to evaluate perspectives of clinical staff on its usability. METHODS: We developed a dashboard based on three principles-"anytime, anywhere, at a glance;" "minimal interruption to workflow;" and "protect patient privacy"-and 3 design features-"geographical layout," "patient-level alert," and "real-time summary data." Items to evaluate the dashboard were selected based on the throughput factor of the conceptual model of ED crowding. Moreover, ED physicians and nurses were surveyed using the system usability scale (SUS) and situation awareness index as well as a questionnaire we created on the basis of the construct of the Situation Awareness Rating Technique. RESULTS: The first version of the ED dashboard was successfully launched in 2013, and it has undergone 3 major revisions since then because of geographical changes in ED and modifications to improve usability. A total of 52 ED staff members participated in the survey. The average SUS score of the dashboard was 67.6 points, which indicates "OK-to-Good" usability. The participants also reported that the dashboard provided efficient "concentration support" (4.15 points), "complexity representation" (4.02 points), "variability representation" (3.96 points), "information quality" (3.94 points), and "familiarity" (3.94 points). However, the "division of attention" was rated at 2.25 points. CONCLUSIONS: We developed a real-time autonomous ED dashboard and successfully used it for 5 years with good evaluation from users.

IVAG: An Integrative Visualization Application for Various Types of Genomic Data Based on R-Shiny and the Docker Platform.

Next-generation sequencing (NGS) technology has become a trend in the genomics research area. There are many software programs and automated pipelines to analyze NGS data, which can ease the pain for traditional scientists who are not familiar with computer programming. However, downstream analyses, such as finding differentially expressed genes or visualizing linkage disequilibrium maps and genome-wide association study (GWAS) data, still remain a challenge. Here, we introduce a dockerized web application written in R using the Shiny platform to visualize pre-analyzed RNA sequencing and GWAS data. In addition, we have integrated a genome browser based on the JBrowse platform and an automated intermediate parsing process required for custom track construction, so that users can easily build and navigate their personal genome tracks with in-house datasets. This application will help scientists perform series of downstream analyses and obtain a more integrative understanding about various types of genomic data by interactively visualizing them with customizable options.

Independent prognostic factors of 861 cases of oral squamous cell carcinoma in Korean adults.

Oral squamous cell carcinoma (OSCC) accounts for 4.5% of all malignant tumors in Korean males and 3.5% in Korean females. The high recurrence rate, and in particular the high local recurrence rate, constitutes a major therapeutic problem for the Korean population, yet there is a paucity of reports addressing the independent predictors of response and survival rate of OSCC in Korea. The present study was designed to investigate the prognostic value of clinical and demographic data within a set of 861 cases of OSCC in Korea. The medical records of 861 OSCC patients who received treatment between 1984 and 1996 at 22 Korean hospitals were reviewed retrospectively with respect to several patient characteristics, including age at diagnosis, gender, location, TNM stage, and treatment. Independent patient-related and treatment-related factors that significantly influenced disease outcome after treatment were analyzed. To assess the independent factors affecting survival rate, univariate and multivariate regression analyses of the survival data were performed using the Cox proportional hazards model. A tree-structured survival model was also derived using survival tree with unbiased detection of interaction (STUDI). The multivariate Cox regression analysis showed that age, gender, composite stage, and treatment method were significant independent prognostic factors. Radiation dose, stage, size of tumor mass, and age of patient also strongly impacted survival time. OSCC is an extremely malignant carcinoma whose prognostic factors are multiple and complex. Based on the findings of this study, we believe that the prognosis of OSCC might depend directly on cancer stage as determined by the TNM system. Furthermore, the survival rate is positively affected by treatment of the neck upon presentation of the cancer, as this can prevent late neck disease due to persistent growth of occult metastases.

Therapeutic efficacy of transcatheter arterial chemoembolization as compared with hepatic resection in hepatocellular carcinoma patients with compensated liver function in a hepatitis B virus-endemic area: a prospective cohort study.

PURPOSE: Identifying a special subgroup of hepatocellular carcinoma (HCC) patients who may benefit from transcatheter arterial chemoembolization (TACE) when compared with the standard treatment of hepatic resection (HR) warrants research in Asian countries. PATIENTS AND METHODS: From January 1993 to December 1994, 182 patients with operable HCC (Child-Pugh class A and International Union Against Cancer [UICC] stage T1-3N0M0) were enrolled. After initial TACE and lipiodol computed tomography, 91 received HR and 91, who refused the operation, received repeated sessions of TACE. After stratification according to the tumor stage (UICC and Cancer of the Liver Italian Program [CLIP]) and lipiodol retention pattern, the survival rates of the two treatment groups were compared. The median follow-up period was 83 months. RESULTS: As of December 31, 2000, 48 patients who underwent HR and 68 patients who underwent TACE had died. In a subgroup analysis according to tumor stage, the HR group survival rate was significantly higher than the TACE group in both UICC T1-2N0M0 (P =.0058) and CLIP 0 (P =.0027) subgroups. However, there was no significant difference in either UICC T3N0M0 (P =.7512) or CLIP 1-2 (P =.5366) subgroups. Even in patients with UICC T1-2N0M0 HCC, when lipiodol was compactly retained, the survival rate of the HR group was comparable to that of the TACE group (P =.0596). CONCLUSION: TACE proved to be as effective as HR in the subpopulations with UICC T3N0M0 or CLIP 1-2 HCC and adequate liver function, and even with UICC T1-2N0M0 HCC when lipiodol was compactly retained in the tumor. In such cases, the choice of treatment modality between TACE and HR may be left to the patient's preference.