Topological data analysis in biomedicine: A review

Significant technological advances made in recent years have shepherded a dramatic increase in utilization of digital technologies for biomedicine- everything from the widespread use of electronic health records to improved medical imaging capabilities and the rising ubiquity of genomic sequencing contribute to a "digitization" of biomedical research and clinical care. With this shift toward computerized tools comes a dramatic increase in the amount of available data, and current tools for data analysis capable of extracting meaningful knowledge from this wealth of information have yet to catch up. This article seeks to provide an overview of emerging mathematical methods with the potential to improve the abilities of clinicians and researchers to analyze biomedical data, but may be hindered from doing so by a lack of conceptual accessibility and awareness in the life sciences research community. In particular, we focus on topological data analysis (TDA), a set of methods grounded in the mathematical field of algebraic topology that seeks to describe and harness features related to the "shape" of data. We aim to make such techniques more approachable to non-mathematicians by providing a conceptual discussion of their theoretical foundations followed by a survey of their published applications to scientific research. Finally, we discuss the limitations of these methods and suggest potential avenues for future work integrating mathematical tools into clinical care and biomedical informatics.

Evaluating the Increased Burden of Cardiorespiratory Illness Visits to Adult Emergency Departments During Flu and Bronchiolitis Outbreaks in the Pediatric Population: Retrospective Multicentric Time Series Analysis

BACKGROUND

Cardiorespiratory decompensation (CRD) visits have a profound effect on adult emergency departments (EDs). Respiratory pathogens like respiratory syncytial virus (RSV) and influenza virus are common reasons for increased activity in pediatric EDs and are associated with CRD in the adult population. Given the seasonal aspects of such challenging pathology, it would be advantageous to predict their variations.

OBJECTIVE

The goal of this study was to evaluate the increased burden of CRD in adult EDs during flu and bronchiolitis outbreaks in the pediatric population.

METHODS

An ecological study was conducted, based on admissions to the adult ED of the Centre Hospitalier Universitaire (CHU) of Grenoble and Saint Etienne from June 29, 2015 to March 22, 2020. The outbreak periods for bronchiolitis and flu in the pediatric population were defined with a decision-making support tool, PREDAFLU, used in the pediatric ED. A Kruskal-Wallis variance analysis and a Spearman monotone dependency were performed in order to study the relationship between the number of adult ED admissions for the International Classification of Diseases (ICD)-10 codes related to cardiorespiratory diagnoses and the presence of an epidemic outbreak as defined with PREDAFLU.

RESULTS

The increase in visits to the adult ED for CRD and the bronchiolitis and flu outbreaks had a similar distribution pattern (CHU Saint Etienne: χ²₃=102.7, P<.001; CHU Grenoble: χ²₃=126.67, P<.001) and were quite dependent in both hospital settings (CHU Saint Etienne: Spearman ρ=0.64; CHU Grenoble: Spearman ρ=0.71). The increase in ED occupancy for these pathologies was also significantly related to the pediatric respiratory infection outbreaks. These 2 criteria gave an idea of the increased workload in the ED due to CRD during the bronchiolitis and flu outbreaks in the pediatric population.

CONCLUSIONS

This study established that CRD visits and bed occupancy for adult EDs were significantly increased during bronchiolitis and pediatric influenza outbreaks. Therefore, a prediction tool for these outbreaks such as PREDAFLU can be used to provide early warnings of increased activity in adult EDs for CRD visits.

Quantifying the Impact of Resuscitation-Team Activation in Hospital Emergency Departments

Hospital emergency department (ED) operations are affected when critically ill or injured patients arrive. Such events often lead to the initiation of specific protocols, referred to as Resuscitation-team Activation (RA), in the ED of Mayo Clinic, Rochester, MN where this study was conducted. RA events lead to the diversion of resources from other patients in the ED to provide care to critically ill patients; therefore, it has an impact on the entire ED system. This paper presents a data-driven and flexible statistical learning model to quantify the impact of RA on the ED. The model learns the pattern of operations in the ED from historical patient arrival and departure timestamps and quantifies the impact of RA by measuring the deviation of the departure of patients during RA from normal processes. The proposed method significantly outperforms baseline methods based on measuring the average time patients spend in the ED.

Emergency department disposition prediction using a deep neural network with integrated clinical narratives and structured data

BACKGROUND

Emergency department (ED) overcrowding has been a serious issue and demands effective clinical decision-making of patient disposition. In previous studies, emergency clinical narratives provide a rich context for clinical decisions. We aimed to develop the disposition prediction model using deep learning modeling strategy with the heterogeneous data, including the physicians' narratives.

METHODS

We constructed a retrospective cohort of all 104,083 ED visits of non-trauma adults during 2017-18 from an academically affiliated ED in Taiwan. 18,308 visits were excluded based on the completeness of each record and the unpredictable dispositions, such as out-of-hospital cardiac arrest, against-advice discharge, and escapes. We integrated subjective section of the first physicians' clinical narratives and structured data (e.g., demographics, triage vital signs, etc.) as available predictors at the first physician-patient encounter. To predict final patient disposition (i.e., hospitalization or discharge), a deep neural network (DNN) model was developed with word embedding, a common natural language processing method. We compared the proposed model to a reference model using the Rapid Emergency Medicine Score, a logistic regression model with structured data, and a DNN model with paragraph vectors. F1 score was used to measure the predictive performance for each model.

RESULTS

The F1 score (with 95 % CI) for the proposed model, the reference model, the logistic regression model with structured data, and the DNN model with paragraph vectors were 0.674 (0.669-0.679), 0.474 (0.469-0.479), 0.547 (0.543-0.551), and 0.602 (0.596-0.607), respectively. While analyzing the relationship between context length and predictive performance under the proposed model, the F1 score at 95th percentile of the word counts was higher than that at 25th percentile of the word counts in chief complaint [0.634 (0.629-0.640) vs. 0.624 (0.620-0.628)] and in present illness [0.671 (0.667-0.674) vs. 0.654 (0.651-0.658)], but not in past medical history [0.674 (0.669-0.679) vs. 0.673 (0.666-0.679)].

CONCLUSIONS

The proposed deep learning model with the usage of the first physicians' clinical narratives and structured data based on natural language processing outperformed the commonly used ones in terms of F1 score. It also evidenced the importance of the subjective section of clinical narratives, which serve as vital predictors for ED clinical decision-making.