Learning hidden patterns from patient multivariate time series data using convolutional neural networks: A case study of healthcare cost prediction

Predicting hospitalization costs for pulmonary tuberculosis patients based on machine learning

BACKGROUND

Pulmonary tuberculosis (PTB) is a prevalent chronic disease associated with a significant economic burden on patients. Using machine learning to predict hospitalization costs can allocate medical resources effectively and optimize the cost structure rationally, so as to control the hospitalization costs of patients better.

METHODS

This research analyzed data (2020-2022) from a Kashgar pulmonary hospital's information system, involving 9570 eligible PTB patients. SPSS 26.0 was used for multiple regression analysis, while Python 3.7 was used for random forest regression (RFR) and MLP. The training set included data from 2020 and 2021, while the test set included data from 2022. The models predicted seven various costs related to PTB patients, including diagnostic cost, medical service cost, material cost, treatment cost, drug cost, other cost, and total hospitalization cost. The model's predictive performance was evaluated using R-square (R2), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE) metrics.

RESULTS

Among the 9570 PTB patients included in the study, the median and quartile of total hospitalization cost were 13,150.45 (9891.34, 19,648.48) yuan. Nine factors, including age, marital status, admission condition, length of hospital stay, initial treatment, presence of other diseases, transfer, drug resistance, and admission department, significantly influenced hospitalization costs for PTB patients. Overall, MLP demonstrated superior performance in most cost predictions, outperforming RFR and multiple regression; The performance of RFR is between MLP and multiple regression; The predictive performance of multiple regression is the lowest, but it shows the best results for Other costs.

CONCLUSION

The MLP can effectively leverage patient information and accurately predict various hospitalization costs, achieving a rationalized structure of hospitalization costs by adjusting higher-cost inpatient items and balancing different cost categories. The insights of this predictive model also hold relevance for research in other medical conditions.

Time Series Prediction Using Deep Learning Methods in Healthcare

Traditional Machine Learning (ML) methods face unique challenges when applied to healthcare predictive analytics. The high-dimensional nature of healthcare data necessitates labor-intensive and time-consuming processes when selecting an appropriate set of features for each new task. Furthermore, ML methods depend heavily on feature engineering to capture the sequential nature of patient data, oftentimes failing to adequately leverage the temporal patterns of medical events and their dependencies. In contrast, recent Deep Learning (DL) methods have shown promising performance for various healthcare prediction tasks by specifically addressing the high-dimensional and temporal challenges of medical data. DL techniques excel at learning useful representations of medical concepts and patient clinical data as well as their nonlinear interactions from high-dimensional raw or minimally-processed healthcare data.

In this paper we systematically reviewed research works that focused on advancing deep neural networks to leverage patient structured time series data for healthcare prediction tasks. To identify relevant studies, we searched MEDLINE, IEEE, Scopus, and ACM digital library for relevant publications through November 4 th , 2021. Overall, we found that researchers have contributed to deep time series prediction literature in ten identifiable research streams: DL models, missing value handling, addressing temporal irregularity, patient representation, static data inclusion, attention mechanisms, interpretation, incorporation of medical ontologies, learning strategies, and scalability. This study summarizes research insights from these literature streams, identifies several critical research gaps, and suggests future research opportunities for DL applications using patient time series data.

Improving Performance of Outcome Prediction for In-patients with Acute Myocardial Infarction Based on Embedding Representation Learned from Electronic Medical Records: Development and Validation Study (Preprint)

Background

The widespread secondary use of electronic medical records (EMRs) promotes health care quality improvement. Representation learning that can automatically extract hidden information from EMR data has gained increasing attention.

Objective

We aimed to propose a patient representation with more feature associations and task-specific feature importance to improve the outcome prediction performance for inpatients with acute myocardial infarction (AMI).

Methods

Medical concepts, including patients’ age, gender, disease diagnoses, laboratory tests, structured radiological features, procedures, and medications, were first embedded into real-value vectors using the improved skip-gram algorithm, where concepts in the context windows were selected by feature association strengths measured by association rule confidence. Then, each patient was represented as the sum of the feature embeddings weighted by the task-specific feature importance, which was applied to facilitate predictive model prediction from global and local perspectives. We finally applied the proposed patient representation into mortality risk prediction for 3010 and 1671 AMI inpatients from a public data set and a private data set, respectively, and compared it with several reference representation methods in terms of the area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), and F1-score.

Results

Compared with the reference methods, the proposed embedding-based representation showed consistently superior predictive performance on the 2 data sets, achieving mean AUROCs of 0.878 and 0.973, AUPRCs of 0.220 and 0.505, and F1-scores of 0.376 and 0.674 for the public and private data sets, respectively, while the greatest AUROCs, AUPRCs, and F1-scores among the reference methods were 0.847 and 0.939, 0.196 and 0.283, and 0.344 and 0.361 for the public and private data sets, respectively. Feature importance integrated in patient representation reflected features that were also critical in prediction tasks and clinical practice.

Conclusions

The introduction of feature associations and feature importance facilitated an effective patient representation and contributed to prediction performance improvement and model interpretation.

Clouds Proportionate Medical Data Stream Analytics for Internet of Things-based Healthcare Systems

Internet of Things (IoT) assisted healthcare systems are designed for providing ubiquitous access and recommendations for personal and distributed electronic health services. The heterogeneous IoT platform assists healthcare services with reliable data management through dedicated computing devices. Healthcare services' reliability depends upon the efficient handling of heterogeneous data streams due to variations and errors. A Proportionate Data Analytics (PDA) for heterogeneous healthcare data stream processing is introduced in this manuscript. This analytics method differentiates the data streams based on variations and errors for satisfying the service responses. The classification is streamlined using linear regression for segregating errors from the variations in different time intervals. The time intervals are differentiated recurrently after detecting errors in the stream's variation. This process of differentiation and classification retains a high response ratio for healthcare services through spontaneous regressions. The proposed method's performance is analyzed using the metrics accuracy, identification ratio, delivery, variation factor, and processing time.

Forecasting Teleconsultation Demand Using an Ensemble CNN Attention-Based BILSTM Model with Additional Variables

To enhance the forecasting accuracy of daily teleconsultation demand, this study proposes an ensemble hybrid deep learning model. The proposed ensemble CNN attention-based BILSTM model (ECA-BILSTM) combines shallow convolutional neural networks (CNNs), attention mechanisms, and bidirectional long short-term memory (BILSTM). Moreover, additional variables are selected according to the characteristics of teleconsultation demand and added to the inputs of forecasting models. To verify the superiority of ECA-BILSTM and the effectiveness of additional variables, two actual teleconsultation datasets collected in the National Telemedicine Center of China (NTCC) are used as the experimental data. Results showed that ECA-BILSTMs can significantly outperform corresponding benchmark models. And two key additional variables were identified for teleconsultation demand prediction improvement. Overall, the proposed ECA-BILSTM model with effective additional variables is a feasible promising approach in teleconsultation demand forecasting.

Early prediction of diagnostic-related groups and estimation of hospital cost by processing clinical notes

As healthcare providers receive fixed amounts of reimbursement for given services under DRG (Diagnosis-Related Groups) payment, DRG codes are valuable for cost monitoring and resource allocation. However, coding is typically performed retrospectively post-discharge. We seek to predict DRGs and DRG-based case mix index (CMI) at early inpatient admission using routine clinical text to estimate hospital cost in an acute setting. We examined a deep learning-based natural language processing (NLP) model to automatically predict per-episode DRGs and corresponding cost-reflecting weights on two cohorts (paid under Medicare Severity (MS) DRG or All Patient Refined (APR) DRG), without human coding efforts. It achieved macro-averaged area under the receiver operating characteristic curve (AUC) scores of 0·871 (SD 0·011) on MS-DRG and 0·884 (0·003) on APR-DRG in fivefold cross-validation experiments on the first day of ICU admission. When extended to simulated patient populations to estimate average cost-reflecting weights, the model increased its accuracy over time and obtained absolute CMI error of 2·40 (1·07%) and 12·79% (2·31%), respectively on the first day. As the model could adapt to variations in admission time, cohort size, and requires no extra manual coding efforts, it shows potential to help estimating costs for active patients to support better operational decision-making in hospitals.

Does Last Year's Cost Predict the Present Cost? An Application of Machine Leaning for the Japanese Area-Basis Public Health Insurance Database

The increasing healthcare cost imposes a large economic burden for the Japanese government. Predicting the healthcare cost may be a useful tool for policy making. A database of the area-basis public health insurance of one city was analyzed to predict the medical healthcare cost by the dental healthcare cost with a machine learning strategy. The 30,340 subjects who had continued registration of the area-basis public health insurance of Ebina city during April 2017 to September 2018 were analyzed. The sum of the healthcare cost was JPY 13,548,831,930. The per capita healthcare cost was JPY 446,567. The proportion of medical healthcare cost, medication cost, and dental healthcare cost was 78%, 15%, and 7%, respectively. By the results of the neural network model, the medical healthcare cost proportionally depended on the medical healthcare cost of the previous year. The dental healthcare cost of the previous year had a reducing effect on the medical healthcare cost. However, the effect was very small. Oral health may be a risk for chronic diseases. However, when evaluated by the healthcare cost, its effect was very small during the observation period.

A scoping review of transfer learning research on medical image analysis using ImageNet

OBJECTIVE

Employing transfer learning (TL) with convolutional neural networks (CNNs), well-trained on non-medical ImageNet dataset, has shown promising results for medical image analysis in recent years. We aimed to conduct a scoping review to identify these studies and summarize their characteristics in terms of the problem description, input, methodology, and outcome.

MATERIALS AND METHODS

To identify relevant studies, MEDLINE, IEEE, and ACM digital library were searched for studies published between June 1st^, 2012 and January 2nd, 2020. Two investigators independently reviewed articles to determine eligibility and to extract data according to a study protocol defined a priori.

RESULTS

After screening of 8421 articles, 102 met the inclusion criteria. Of 22 anatomical areas, eye (18%), breast (14%), and brain (12%) were the most commonly studied. Data augmentation was performed in 72% of fine-tuning TL studies versus 15% of the feature-extracting TL studies. Inception models were the most commonly used in breast related studies (50%), while VGGNet was the common in eye (44%), skin (50%) and tooth (57%) studies. AlexNet for brain (42%) and DenseNet for lung studies (38%) were the most frequently used models. Inception models were the most frequently used for studies that analyzed ultrasound (55%), endoscopy (57%), and skeletal system X-rays (57%). VGGNet was the most common for fundus (42%) and optical coherence tomography images (50%). AlexNet was the most frequent model for brain MRIs (36%) and breast X-Rays (50%). 35% of the studies compared their model with other well-trained CNN models and 33% of them provided visualization for interpretation.

DISCUSSION

This study identified the most prevalent tracks of implementation in the literature for data preparation, methodology selection and output evaluation for various medical image analysis tasks. Also, we identified several critical research gaps existing in the TL studies on medical image analysis. The findings of this scoping review can be used in future TL studies to guide the selection of appropriate research approaches, as well as identify research gaps and opportunities for innovation.