Development of Machine Learning Prediction Models for Self-Extubation After Delirium Using Emergency Department Data.

Delirium is common in the emergency department, and once it develops, there is a risk of self-extubation of drains and tubes, so it is critical to predict delirium before it occurs. Machine learning was used to create two prediction models in this study: one for predicting the occurrence of delirium and one for predicting self-extubation after delirium. Each model showed high discriminative performance, indicating the possibility of selecting high-risk cases. Visualization of predictors using Shapley additive explanation (SHAP), a machine learning interpretability method, showed that the predictors of delirium were different from those of self-extubation after delirium. Data-driven decisions, rather than empirical decisions, on whether or not to use physical restraints or other actions that cause patient suffering will result in improved value in medical care.

Temporal Generalizability of Machine Learning Models for Predicting Postoperative Delirium Using Electronic Health Record Data: Model Development and Validation Study.

BACKGROUND: Although machine learning models demonstrate significant potential in predicting postoperative delirium, the advantages of their implementation in real-world settings remain unclear and require a comparison with conventional models in practical applications. OBJECTIVE: The objective of this study was to validate the temporal generalizability of decision tree ensemble and sparse linear regression models for predicting delirium after surgery compared with that of the traditional logistic regression model. METHODS: The health record data of patients hospitalized at an advanced emergency and critical care medical center in Kumamoto, Japan, were collected electronically. We developed a decision tree ensemble model using extreme gradient boosting (XGBoost) and a sparse linear regression model using least absolute shrinkage and selection operator (LASSO) regression. To evaluate the predictive performance of the model, we used the area under the receiver operating characteristic curve (AUROC) and the Matthews correlation coefficient (MCC) to measure discrimination and the slope and intercept of the regression between predicted and observed probabilities to measure calibration. The Brier score was evaluated as an overall performance metric. We included 11,863 consecutive patients who underwent surgery with general anesthesia between December 2017 and February 2022. The patients were divided into a derivation cohort before the COVID-19 pandemic and a validation cohort during the COVID-19 pandemic. Postoperative delirium was diagnosed according to the confusion assessment method. RESULTS: A total of 6497 patients (68.5, SD 14.4 years, women n=2627, 40.4%) were included in the derivation cohort, and 5366 patients (67.8, SD 14.6 years, women n=2105, 39.2%) were included in the validation cohort. Regarding discrimination, the XGBoost model (AUROC 0.87-0.90 and MCC 0.34-0.44) did not significantly outperform the LASSO model (AUROC 0.86-0.89 and MCC 0.34-0.41). The logistic regression model (AUROC 0.84-0.88, MCC 0.33-0.40, slope 1.01-1.19, intercept -0.16 to 0.06, and Brier score 0.06-0.07), with 8 predictors (age, intensive care unit, neurosurgery, emergency admission, anesthesia time, BMI, blood loss during surgery, and use of an ambulance) achieved good predictive performance. CONCLUSIONS: The XGBoost model did not significantly outperform the LASSO model in predicting postoperative delirium. Furthermore, a parsimonious logistic model with a few important predictors achieved comparable performance to machine learning models in predicting postoperative delirium.

Non-linear association between body weight and functional outcome after acute ischemic stroke.

This study aimed to determine whether body weight is associated with functional outcome after acute ischemic stroke. We measured the body mass index (BMI) and assessed clinical outcomes in patients with acute ischemic stroke. The BMI was categorized into underweight (< 18.5 kg/m2), normal weight (18.5-22.9 kg/m2), overweight (23.0-24.9 kg/m2), and obesity (≥ 25.0 kg/m2). The association between BMI and a poor functional outcome (modified Rankin Scale [mRS] score: 3-6) was evaluated. We included 11,749 patients with acute ischemic stroke (70.3 ± 12.2 years, 36.1% women). The risk of a 3-month poor functional outcome was higher for underweight, lower for overweight, and did not change for obesity in reference to a normal weight even after adjusting for covariates by logistic regression analysis. Restricted cubic splines and SHapley Additive exPlanation values in eXtreme Gradient Boosting model also showed non-linear relationships. Associations between BMI and a poor functional outcome were maintained even after excluding death (mRS score: 3-5) or including mild disability (mRS score: 2-6) as the outcome. The associations were strong in older patients, non-diabetic patients, and patients with mild stroke. Body weight has a non-linear relationship with the risk of a poor functional outcome after acute ischemic stroke.

Communications Through Contemporary Tools of Information and Communication Technology: Cross-sectional Study Evaluating Health Among Separated Family Members.

BACKGROUND: The number of single-living workers separated from their spouses and families has been increasing due to the need to create a balance between life and work. Workers are assigned everywhere in globalized workplaces while also caring for their family members in the context of Japan's aging society. At the same time, the mental and health status of persons living separately from their families is a matter of concern. The development of interpersonal communication means using information and communications technology (ICT) tools and the internet is remarkable, enabling simultaneous 2-way communication across distances and national borders. The easy accessibility to simultaneous communication is expected to improve the psychosocial status of isolated family members. OBJECTIVE: This study aims to clarify the health benefits of ICT by using a psychosocial health assessment, the characteristics of ICT tools, and the frequency of communication among the workers and their families who live separately. METHODS: This was a cross-sectional study planned and conducted in Japan. Study participants, including adults who live separately from other family members or have separately living family members due to work, were recruited to answer a web response survey about ICT usage status, health status, and life and society evaluation. This study recruited 73 participants divided into 2 groups by their communication tools and frequencies, and their separated life, health, and psychosocial status were statistically compared. RESULTS: Among the 73 study participants, 15 were categorized in the high communication-skilled (HCS) group that used both types of ICT tools to communicate frequently: "live," such as video chat and voice call, and "nonlive," such as SMS text message service and email. A simple comparison between the HCS and reference groups showed significant differences in the cohesion with the neighborhood (P=.03), perceived social position (P=.01), and happiness (P<.001); however, there were no significant differences in the health (psychological distress, P=.08; self-rated health, P=.07), lifestyle (drinking, P>.99; current smoking, P=.37), and dyadic trust in family members living separately (P=.80). Further, in a multivariate regression analysis adjusted for confounding factors, such as educational history, age, gender, and job status, poor subjective health showed a prevalence odds ratio of less than 1 (OR 0.17, 95% CI 0.03-1.02). The HCS group showed significant positive relationships in the cohesion score with the neighborhood (P=.01; ß=2.40, 95% CI 0.56-4.24), perceived social position (P=.03; ß=1.17, 95% CI 0.11-2.23), and happiness score (P=.002; ß=1.46, 95% CI 0.58-2.34) in the same multivariate regression models. CONCLUSIONS: This study suggested that people who frequently communicate with separated family members by taking advantage of various ICT tools can maintain a better mental state and better social relations among those who live alone and are separated from their families.

Explanation of machine learning models using shapley additive explanation and application for real data in hospital.

BACKGROUND AND OBJECTIVE: When using machine learning techniques in decision-making processes, the interpretability of the models is important. In the present paper, we adopted the Shapley additive explanation (SHAP), which is based on fair profit allocation among many stakeholders depending on their contribution, for interpreting a gradient-boosting decision tree model using hospital data. METHODS: For better interpretability, we propose two novel techniques as follows: (1) a new metric of feature importance using SHAP and (2) a technique termed feature packing, which packs multiple similar features into one grouped feature to allow an easier understanding of the model without reconstruction of the model. We then compared the explanation results between the SHAP framework and existing methods using cerebral infarction data from our hospital. RESULTS: The interpretation by SHAP was mostly consistent with that by the existing methods. We showed how the A/G ratio works as an important prognostic factor for cerebral infarction using proposed techniques. CONCLUSION: Our techniques are useful for interpreting machine learning models and can uncover the underlying relationships between features and outcome.

Machine learning for classification of postoperative patient status using standardized medical data.

BACKGROUND AND OBJECTIVE: Real-world evidence is defined as clinical evidence regarding the use and potential benefits or risks of a medical product derived from real-world data analyses. Standardization and structuring of data are necessary to analyze medical real-world data collected from different medical institutions. An electronic message and repository have been developed to link electronic medical records in this research project, which has simplified the data integration. Therefore, this paper proposes an analysis method and learning health systems to determine the priority of clinical intervention by clustering and visualizing time-series and prioritizing patient outcomes and status during hospitalization. METHODS: Common data items for reimbursement (Diagnosis Procedure Combination [DPC]) and clinical pathway data were examined in this project at each participating institution that runs the verification test. Long-term hospitalization data were analyzed using the data stored in the cloud platform of the institutions' repositories using multiple machine learning methods for classification, visualization, and interpretation. RESULTS: The ePath platform contributed to integrate the standardized data from multiple institutions. The distribution of DPC items or variances could be confirmed by clustering, temporal tendency through the directed graph, and extracting variables that contributed to the prediction and evaluation of SHapley Additive Explanation effects. Constipation was determined to be the risk factor most strongly related to long-term hospitalization. Drainage management was identified as a factor that can improve long-term hospitalization. These analyses effectively extracted patient status to provide feedback to the learning health system. CONCLUSIONS: We successfully generated evidence of medical processes by gathering patient status, medical purposes, and patient outcomes with high data quality from multiple institutions, which were difficult with conventional electronic medical records. Regarding the significant analysis results, the learning health system will be used on this project to provide feedback to each institution, operate it for a certain period, and analyze and re-evaluate it.

A functional learning health system in Japan: Experience with processes and information infrastructure toward continuous health improvement.

Introduction and definition of the term Learning Health System (LHS) appears to have occurred initially around 2007. Prior to this and the introduction of electronic health records (EHR), a predecessor could be found in the Clinical Pathways concept as a standard medical care plan and a tool to improve medical quality. Since 1997, Japan's Saiseikai Kumamoto Hospital (SKH) has been studying and implementing Clinical Pathways. In 2010, they implemented EHR, which facilitated the collection of structured data in common templates that aligned with outcome measurements defined through Japan's Society of Clinical Pathways. For each patient at this hospital, variances from the desired outcomes have been recorded, producing volumes of structured data in formats that could readily be aggregated and analyzed. A visualization tool was introduced to display graphs on the home page of the EHR such that each patient can be compared to similar patients. Knowledge learned from patient care is shared regularly through Clinical Pathways meetings that are supported by all staff within the hospital. The SKH experience over the past two decades is worth exploring further in the context of the development of a fully functional LHS and the attributes/characteristics thereof. In this report, the SKH experience and processes are compared with previously published attributes of a fully functional LHS (ie, characteristics of an LHS that can indicate maturity). Specific examples of the SKH system are detailed with respect to leveraging knowledge gained to change performance that improves patient care as prescribed by learning health cycles. The SKH experience and its information infrastructure and culture exemplify a functional LHS, which is now being expanded to additional hospitals with the hope that it can be scaled and serve as a solid platform for measures aimed at improving medical care, thus establishing broader and more global learning health systems.

Association of serum bilirubin levels with risk of cancer development and total death.

Serum levels of bilirubin, a strong antioxidant, may influence cancer risk. We aimed to assess the association between serum bilirubin levels and cancer risk. Data were retrieved from 10-year electronic medical records at Kyushu University Hospital (Japan) for patients aged 20 to 69 years old. The associations of baseline bilirubin levels with cancer risk (lung, colon, breast, prostate, and cervical) were evaluated using a gradient boosting decision tree (GBDT) model, a machine learning algorithm, and Cox proportional hazard regression model, adjusted for age, smoking, body mass index, and diabetes. The number of study subjects was 29,080. Median follow-up time was 4.7 years. GBDT models illustrated that baseline bilirubin levels were negatively and non-linearly associated with the risk of lung (men), colon, and cervical cancer. In contrast, a U-shaped association was observed for breast and prostate cancer. Cox hazard regression analyses confirmed that baseline bilirubin levels (< 1.2 mg/dL) were negatively associated with lung cancer risk in men (HR = 0.474, 95% CI 0.271-0.828, P = 0.009) and cervical cancer risk (HR = 0.365, 95% CI 0.136-0.977, P = 0.045). Additionally, low bilirubin levels (< 0.6 mg/dL) were associated with total death (HR = 1.744, 95% CI 1.369-2.222, P < 0.001). Serum bilirubin may have a beneficial effect on the risk of some types of cancers.

Portable health clinic for sustainable care of mothers and newborns in rural Bangladesh.

BACKGROUND AND OBJECTIVE: Regular health checkups are important for mothers and newborns to detect health problems at an early stage; however, this is often difficult in resource-limited settings. Therefore, the portable health clinic (PHC) for maternal and child health (MCH), a telemedicine health checkup system, was introduced as an intervention study in a rural area in Bangladesh. The aim of this research project was to report findings that we had observed at a mid-point of the intervention period. METHODS: This was an intervention study conducted in Shariatpur, Bangladesh. The study population included pregnant/parturient women aged 15-49 years and their newborns. With the help of the newly created PHC for MCH, health workers, with a set of sensor devices in an attaché case, visited mothers and newborns at home to examine their health status. Their health status was triaged into four categories using a data management application, and in cases of affected or emergent health status, they were placed on remote video consultation with a doctor. RESULTS: In total, 94 women were included in the PHC for MCH intervention. The rate of participants who received antenatal care at least four times or postnatal care at least once increased (from 29% to 51%, and from 27% to 78%, respectively) compared with before introducing PHC for MCH. Using the PHC for MCH, we detected health problems in pregnant/parturient women; a relatively high percentage had anemia (45-54%) and/or abnormal pulse rate (20-40%). Moreover, after introducing the PHC for MCH, more than 40% of women who received multiple antenatal care or postnatal care checkups improved their health status. CONCLUSIONS: The PHC for MCH could be an effective system to improve the health of mothers and newborns by increasing the availability of care. In the future, this system is expected to be used as a primary resource for maternity healthcare, not only in rural areas but also in other social environments.

Impact of a learning health system on acute care and medical complications after intracerebral hemorrhage.

INTRODUCTION: Patients with stroke often experience pneumonia during the acute stage after stroke onset. Oral care may be effective in reducing the risk of stroke-associated pneumonia (SAP). We aimed to determine the changes in oral care, as well as the incidence of SAP, in patients with intracerebral hemorrhage, following implementation of a learning health system in our hospital. METHODS: We retrospectively analyzed the data of 1716 patients with intracerebral hemorrhage who were hospitalized at a single stroke center in Japan between January 2012 and December 2018. Data were stratified on the basis of three periods of evolving oral care: period A, during which conventional, empirically driven oral care was provided (n = 725); period B, during which standardized oral care was introduced, with SAP prophylaxis based on known risk factors (n = 469); and period C, during which oral care was risk-appropriate based on learning health system data (n = 522). Logistic regression analysis was performed to evaluate associations between each of the three treatment approaches and the risk of SAP. RESULTS: Among the included patients, the mean age was 71.3 ± 13.6 years; 52.6% of patients were men. During the course of each period, the frequency of oral care within 24 hours of admission increased (P < .001), as did the adherence rate to oral care ≥3 times per day (P < .001). After adjustment for confounding factors, a change in the risk of SAP was not observed in period B; however, the risk significantly decreased in period C (odds ratio 0.61; 95% confidence interval 0.43-0.87) compared with period A. These associations were maintained for SAP diagnosed using strict clinical criteria or after exclusion of 174 patients who underwent neurosurgical treatment. CONCLUSIONS: Risk-appropriate care informed by the use of learning health system data could improve care and potentially reduce the risk of SAP in patients with intracerebral hemorrhage in the acute stage.

Effect of treatment modality and cerebral vasospasm agent on patient outcomes after aneurysmal subarachnoid hemorrhage in the elderly aged 75 years and older.

OBJECTIVE: We sought to examine whether the effect of treatment modality and drugs for cerebral vasospasm on clinical outcomes differs between elderly and non-elderly subarachnoid hemorrhage (SAH) patients in Japan. METHODS: We analyzed the J-ASPECT Study Diagnosis Procedure Combination database (n = 17,343) that underwent clipping or coiling between 2010 and 2014 in 579 hospitals. We stratified patients into two groups according to their age (elderly [≥75 years old], n = 3,885; non-elderly, n = 13,458). We analyzed the effect of treatment modality and anti-vasospasm agents (fasudil hydrochloride, ozagrel sodium, cilostazol, statin, eicosapentaenoic acid [EPA], and edaravone) on in-hospital poor outcomes (mRS 3-6 at discharge) and mortality using multivariable analysis. RESULTS: The elderly patients were more likely to be female, have impaired levels of consciousness and comorbidity, and less likely to be treated with clipping and anti-vasospasm agents, except for ozagrel sodium and statin. In-hospital mortality and poor outcomes were higher in the elderly (15.8% vs. 8.5%, 71.7% vs. 36.5%). Coiling was associated with higher mortality (odds ratio 1.43, 95% confidence interval 1.2-1.7) despite a lower proportion of poor outcomes (0.84, 0.75-0.94) in the non-elderly, in contrast to no effect on clinical outcomes in the elderly. A comparable effect of anti-vasospasm agents on mortality was observed between non-elderly and elderly for fasudil hydrochloride (non-elderly: 0.20, 0.17-0.24), statin (0.63, 0.50-0.79), ozagrel sodium (0.72, 0.60-0.86), and cilostazol (0.63, 0.51-0.77). Poor outcomes were inversely associated with fasudil hydrochloride (0.59, 0.51-0.68), statin (0.84, 0.75-0.94), and EPA (0.83, 0.72-0.94) use in the non-elderly. No effect of these agents on poor outcomes was observed in the elderly. CONCLUSIONS: In contrast to the non-elderly, no effect of treatment modality on clinical outcomes were observed in the elderly. A comparable effect of anti-vasospasm agents was observed on mortality, but not on functional outcomes, between the non-elderly and elderly.

Stroke Prognostic Scores and Data-Driven Prediction of Clinical Outcomes After Acute Ischemic Stroke.

Background and Purpose- Several stroke prognostic scores have been developed to predict clinical outcomes after stroke. This study aimed to develop and validate novel data-driven predictive models for clinical outcomes by referring to previous prognostic scores in patients with acute ischemic stroke in a real-world setting. Methods- We used retrospective data of 4237 patients with acute ischemic stroke who were hospitalized in a single stroke center in Japan between January 2012 and August 2017. We first validated point-based stroke prognostic scores (preadmission comorbidities, level of consciousness, age, and neurological deficit [PLAN] score, ischemic stroke predictive risk score [IScore], and acute stroke registry and analysis of Lausanne [ASTRAL] score in all patients; Houston intraarterial recanalization therapy [HIAT] score, totaled health risks in vascular events [THRIVE] score, and stroke prognostication using age and National Institutes of Health Stroke Scale-100 [SPAN-100] in patients who received reperfusion therapy) in our cohort. We then developed predictive models using all available data by linear regression or decision tree ensembles (random forest and gradient boosting decision tree) and evaluated their area under the receiver operating characteristic curve for clinical outcomes after repeated random splits. Results- The mean (SD) age of the patients was 74.7 (12.9) years and 58.3% were men. Area under the receiver operating characteristic curves (95% CIs) of prognostic scores in our cohort were 0.92 PLAN score (0.90-0.93), 0.86 for IScore (0.85-0.87), 0.85 for ASTRAL score (0.83-0.86), 0.69 for HIAT score (0.62-0.75), 0.70 for THRIVE score (0.64-0.76), and 0.70 for SPAN-100 (0.63-0.76) for poor functional outcomes, and 0.87 for PLAN score (0.85-0.90), 0.88 for IScore (0.86-0.91), and 0.88 ASTRAL score (0.85-0.91) for in-hospital mortality. Internal validation of data-driven prediction models showed that their area under the receiver operating characteristic curves ranged between 0.88 and 0.94 for poor functional outcomes and between 0.84 and 0.88 for in-hospital mortality. Ensemble models of a decision tree tended to outperform linear regression models in predicting poor functional outcomes but not in predicting in-hospital mortality. Conclusions- Stroke prognostic scores perform well in predicting clinical outcomes after stroke. Data-driven models may be an alternative tool for predicting poststroke clinical outcomes in a real-world setting.

Portable Health Clinic: An Advanced Tele-Healthcare System for Unreached Communities.

The Portable Health Clinic (PHC) system endeavors to take healthcare facilities along with remote doctors' consultancy to the doorsteps of the unreached people using an advanced telemedicine system. Thus, the necessity of having physical healthcare peripheries specially in the developing countries can be mitigated. The PHC system promotes preventive healthcare by encouraging regular health checkups so that diseases can be prevented as well as their severity can be mitigated, leading to a reduction on healthcare expenses. Thus, the number of patients along with excessive workload on existing healthcare human resources can be minimized. The current project in rural Bangladesh alone has served more than 41,000 people so far by the PHC system and a simple analysis of this data shows some significant findings on regional health status. A simple expansion of this program, covering a wider service area, can produce a big data to reflect the whole country's health profile.

The Development of an Electronic Phenotyping Algorithm for Identifying Rhabdomyolysis Patients in the MID-NET Database.

We aimed to develop rhabdomyolysis (RB) phenotyping algorithms using machine learning techniques and to create subphenotyping algorithms to identify RB patients who lack RB diagnosis. Two pattern algorithms, one with a focus on improving predictive value and one focused on improving sensitivity, were finally created and had a high area under the curve value of 0.846. Although we were unable to create subphenotyping algorithms, an attempt to detect unknown RB patients is important for epidemiological studies.

Effect of Governance Functionality for Data Standardization Management of the Medical Information Database Network Project.

Data standardiztion an important aspect to ensure data quality for utilizing large-scale, medical information databases such as the Medical Information Database Network (MID-NET) Project in Japan. We established a governance center to assess the consistency of standard codes across MID-NET-cooperating medical institutions. Moreover, we developed a real-time validation tool and determined its effect in improving data quality in medical institutions by providing a central feedback on the detected differences in standard disease-name codes.

Interpretable Machine Learning Techniques for Causal Inference Using Balancing Scores as Meta-features.

Estimating individual causal effect is important for decision making in many fields especially for medical interventions. We propose an interpretable and accurate algorithm for estimating causal effects from observational data. The proposed scheme is combining multiple predictors' outputs by an interpretable predictor such as linear predictor and if then rules. We secure interpretability using the interpretable predictor and balancing scores in causal inference studies as meta-features. For securing accuracy, we adapt machine learning algorithms for calculating balancing scores. We analyze the effect of t-PA therapy for stroke patients using real-world data, which has 64,609 records with 362 variables and interpret results. The results show that cross validation AUC of the proposed scheme is little less than original machine learning scheme; however, the proposed scheme provides interpretability that t-PA therapy is effective for severe patients.

Postnatal care could be the key to improving the continuum of care in maternal and child health in Ratanakiri, Cambodia.

In South-East Asia, the maternal and child mortality rate has declined over the past decades; however, it varies among and within the countries in the region, including Cambodia. The continuum of care is an integrated series of care that women and children are required to avail continuously from pregnancy to the child/motherhood period. This study aimed to assess the completion rate of the continuum of care and examine the factors associated with the continuum of care in Ratanakiri, Cambodia. A cross-sectional study was conducted in Ratanakiri. Overall, 377 women were included, and data were collected via face-to-face interviews using a semi-structured questionnaire. Among them, 5.0% completed the continuum of care (antenatal care at least four times, delivery by skilled birth attendant, and postnatal care at least once). Meanwhile, 18.8% did not receive any care during pregnancy, delivery, and after birth. The highest discontinuation rate was at the postnatal care stage (73.6%). Not receiving any perinatal care was associated with neonatal complications at 6 weeks after birth (adjusted odds ratio [AOR]: 3.075; 95% confidence interval [CI]: 1.310-7.215). Furthermore, a long distance to the health center was negatively associated with completion of the continuum of care (AOR: 0.877; 95% CI: 0.791-0.972). This study indicates the need for efforts to reduce the number of women who discontinue from the continuum of care, as well as who do not receive any care to avoid neonatal complications. Since the discontinuation rate was highest at the postnatal care, postnatal care needs to be promoted more through the antenatal care and delivery services. Furthermore, given that long distance to health facilities was a barrier for receiving the care continuously, our findings suggest the need for a village-based health care system that can provide the basic continuum of care in remote areas.

Prediction and Factor Extraction of Drug Function by Analyzing Medical Records in Developing Countries.

The World Health Organization has declared Bangladesh one of 58 countries facing acute Human Resources for Health (HRH) crisis. Artificial intelligence in healthcare has been shown to be successful for diagnostics. Using machine learning to predict pharmaceutical prescriptions may solve HRH crises. In this study, we investigate a predictive model by analyzing prescription data of 4,543 subjects in Bangladesh. We predict the function of prescribed drugs, comparing three machine-learning approaches. The approaches compare whether a subject shall be prescribed medicine from the 21 most frequently prescribed drug functions. Receiver Operating Characteristics (ROC) were selected as a way to evaluate and assess prediction models. The results show the drug function with the best prediction performance was oral hypoglycemic drugs, which has an average AUC of 0.962. To understand how the variables affect prediction, we conducted factor analysis based on tree-based algorithms and natural language processing techniques.

Extracting Predictive Indicator for Prognosis of Cerebral Infarction Using Machine Learning Techniques.

Identifying important predicative indicators for prognosis is useful since these factors help for understanding diseases and determining treatments for patients. We extracted important factors for prognosis of cerebral infarction from EHR. We analyzed EHR data of 1,697 patients with 1,602 variables using gradient boosting decision tree. Extracted factors include not only well-known factors such as NIHSS but also new factors such as albumin-globulin ratio.

Mobile Healthcare System for Health Checkups and Telemedicine in Post-Disaster Situations.

Portable Healthcare Clinic (PHC) is a mobile healthcare system comprising of medical sensors and health assessment criteria. It has been applied in Bangladesh for the last two years as a pilot program to identify non-communicable diseases. In this study, we adapted PHC to fit post-disaster conditions. The PHC health assessment criteria are redesigned to deal with emergency cases and healthcare worker insufficiency. A new algorithm makes an initial assessment of age, symptoms, and whether the person is seeing a doctor. These changes will make the turn-around time shorter and will enable reaching the most affected patients better. We tested the operability and turn-around time of the adapted system at the debris flow disaster shelters in Hiroshima, Japan. Changing the PHC health assessment criteria and other solutions such as a list of medicine preparation makes the PHC system switch into an emergency mode more smoothly following a natural disaster.