Machine Learning for Health Informatics. In: Holzinger A, editor. Machine Learning for Health Informatics. Cham: Springer International Publishing; 2016. pp. 1-24. [DOI: 10.1007/978-3-319-50478-0_1]

Big Data Analytics to Reduce Preventable Hospitalizations-Using Real-World Data to Predict Ambulatory Care-Sensitive Conditions

The purpose of this study was to develop a prediction model to identify individuals and populations with a high risk of being hospitalized due to an ambulatory care-sensitive condition who might benefit from preventative actions or tailored treatment options to avoid subsequent hospital admission. A rate of 4.8% of all individuals observed had an ambulatory care-sensitive hospitalization in 2019 and 6389.3 hospital cases per 100,000 individuals could be observed. Based on real-world claims data, the predictive performance was compared between a machine learning model (Random Forest) and a statistical logistic regression model. One result was that both models achieve a generally comparable performance with c-values above 0.75, whereas the Random Forest model reached slightly higher c-values. The prediction models developed in this study reached c-values comparable to existing study results of prediction models for (avoidable) hospitalization from the literature. The prediction models were designed in such a way that they can support integrated care or public and population health interventions with little effort with an additional risk assessment tool in the case of availability of claims data. For the regions analyzed, the logistic regression revealed that switching to a higher age class or to a higher level of long-term care and unit from prior hospitalizations (all-cause and due to an ambulatory care-sensitive condition) increases the odds of having an ambulatory care-sensitive hospitalization in the upcoming year. This is also true for patients with prior diagnoses from the diagnosis groups of maternal disorders related to pregnancy, mental disorders due to alcohol/opioids, alcoholic liver disease and certain diseases of the circulatory system. Further model refinement activities and the integration of additional data, such as behavioral, social or environmental data would improve both model performance and the individual risk scores. The implementation of risk scores identifying populations potentially benefitting from public health and population health activities would be the next step to enable an evaluation of whether ambulatory care-sensitive hospitalizations can be prevented.

Computer Based Diagnosis of Some Chronic Diseases: A Medical Journey of the Last Two Decades

Disease prediction from diagnostic reports and pathological images using artificial intelligence (AI) and machine learning (ML) is one of the fastest emerging applications in recent days. Researchers are striving to achieve near-perfect results using advanced hardware technologies in amalgamation with AI and ML based approaches. As a result, a large number of AI and ML based methods are found in the literature. A systematic survey describing the state-of-the-art disease prediction methods, specifically chronic disease prediction algorithms, will provide a clear idea about the recent models developed in this field. This will also help the researchers to identify the research gaps present there. To this end, this paper looks over the approaches in the literature designed for predicting chronic diseases like Breast Cancer, Lung Cancer, Leukemia, Heart Disease, Diabetes, Chronic Kidney Disease and Liver Disease. The advantages and disadvantages of various techniques are thoroughly explained. This paper also presents a detailed performance comparison of different methods. Finally, it concludes the survey by highlighting some future research directions in this field that can be addressed through the forthcoming research attempts.

Context-Based Identification of Muscle Invasion Status in Patients With Bladder Cancer Using Natural Language Processing

PURPOSE

Mortality from bladder cancer (BC) increases exponentially once it invades the muscle, with inherent challenges delineating at the population level. We sought to develop and validate a natural language processing (NLP) model for automatically identifying patients with muscle-invasive bladder cancer (MIBC).

METHODS

All patients with a Current Procedural Terminology code for transurethral resection of bladder tumor (TURBT; n = 76,060) were selected from the Department of Veterans Affairs (VA) database. A sample of 600 patients (with 2,337 full-text notes) who had TURBT and confirmed pathology results were selected for NLP model development and validation. The NLP performance was assessed by calculating the sensitivity, specificity, positive predictive value, negative predictive value, F1 score, and overall accuracy at the individual note and patient levels.

RESULTS

In the validation cohort, the NLP model had average overall accuracies of 94% and 96% at the note and patient levels. Specifically, the F1 score and overall accuracy for predicting muscle invasion at the patient level were 0.87% and 96%, respectively. The model classified nonmuscle-invasive bladder cancer (NMIBC) with overall accuracies of 90% and 93% at the note and patient levels. When applying the model to 71,200 patients VA-wide, the model classified 13,642 (19%) as having MIBC and 47,595 (66%) as NMIBC and was able to identify invasion status for 96% of patients with TURBT at the population level. Inherent limitations include a relatively small training set, given the size of the VA population.

CONCLUSION

This NLP model, with high accuracy, may be a practical tool for efficiently identifying BC invasion status and aid in population-based BC research.

Machine Learning Approaches to Retrieve High-Quality, Clinically Relevant Evidence From the Biomedical Literature: Systematic Review

BACKGROUND

The rapid growth of the biomedical literature makes identifying strong evidence a time-consuming task. Applying machine learning to the process could be a viable solution that limits effort while maintaining accuracy.

OBJECTIVE

The goal of the research was to summarize the nature and comparative performance of machine learning approaches that have been applied to retrieve high-quality evidence for clinical consideration from the biomedical literature.

METHODS

We conducted a systematic review of studies that applied machine learning techniques to identify high-quality clinical articles in the biomedical literature. Multiple databases were searched to July 2020. Extracted data focused on the applied machine learning model, steps in the development of the models, and model performance.

RESULTS

From 3918 retrieved studies, 10 met our inclusion criteria. All followed a supervised machine learning approach and applied, from a limited range of options, a high-quality standard for the training of their model. The results show that machine learning can achieve a sensitivity of 95% while maintaining a high precision of 86%.

CONCLUSIONS

Machine learning approaches perform well in retrieving high-quality clinical studies. Performance may improve by applying more sophisticated approaches such as active learning and unsupervised machine learning approaches.

FeARH: Federated machine learning with anonymous random hybridization on electronic medical records

Electrical medical records are restricted and difficult to centralize for machine learning model training due to privacy and regulatory issues. One solution is to train models in a distributed manner that involves many parties in the process. However, sometimes certain parties are not trustable, and in this project, we aim to propose an alternative method to traditional federated learning with central analyzer in order to conduct training in a situation without a trustable central analyzer. The proposed algorithm is called "federated machine learning with anonymous random hybridization (abbreviated as 'FeARH')", using mainly hybridization algorithm to degenerate the integration of connections between medical record data and models' parameters by adding randomization into the parameter sets shared to other parties. Based on our experiment, our new algorithm has similar AUCROC and AUCPR results compared with machine learning in a centralized manner and original federated machine learning.

Greg, ML – Machine Learning for Healthcare at a Scale

This paper introduces the Greg, ML platform, a machine-learning engine and toolset conceived to generate automatic diagnostic suggestions based on patient profiles. Greg, ML departs from many other experiences in machine learning for healthcare in the fact that it was designed to handle a large number of different diagnoses, in the order of the hundreds. We discuss the architecture that stands at the core of Greg, ML, designed to handle the complex challenges posed by this ambitious goal, and confirm its effectiveness with experimental results based on the working prototype we have developed. Finally, we discuss challenges and opportunities related to the use of this kind of tools in medicine, and some important lessons learned while developing the tool. In this respect, we underline that Greg, ML should be conceived primarily as a support for expert doctors in their diagnostic decisions, and can hardly replace humans in their judgment.

Machine learning predictive system based upon radiodensitometric distributions from mid-thigh CT images

The nonlinear trimodal regression analysis (NTRA) method based on radiodensitometric CT images distributions was developed for the quantitative characterization of soft tissue changes according to the lower extremity function of elderly subjects. In this regard, the NTRA method defines 11 subject-specific soft tissue parameters and has illustrated high sensitivity to changes in skeletal muscle form and function. The present work further explores the use of these 11 NTRA parameters in the construction of a machine learning (ML) system to predict body mass index and isometric leg strength using tree-based regression algorithms. Results obtained from these models demonstrate that when using an ML approach, these soft tissue features have a significant predictive value for these physiological parameters. These results further support the use of NTRA-based ML predictive assessment and support the future investigation of other physiological parameters and comorbidities.

Analytic Considerations for Repeated Measures of eGFR in Cohort Studies of CKD

Repeated measures of various biomarkers provide opportunities for us to enhance understanding of many important clinical aspects of CKD, including patterns of disease progression, rates of kidney function decline under different risk factors, and the degree of heterogeneity in disease manifestations across patients. However, because of unique features, such as correlations across visits and time dependency, these data must be appropriately handled using longitudinal data analysis methods. We provide a general overview of the characteristics of data collected in cohort studies and compare appropriate statistical methods for the analysis of longitudinal exposures and outcomes. We use examples from the Chronic Renal Insufficiency Cohort Study to illustrate these methods. More specifically, we model longitudinal kidney outcomes over annual clinical visits and assess the association with both baseline and longitudinal risk factors.

Introduction to MAchine Learning & Knowledge Extraction (MAKE)

The grand goal of Machine Learning is to develop software which can learn from previous experience—similar to how we humans do. Ultimately, to reach a level of usable intelligence, we need (1) to learn from prior data, (2) to extract knowledge, (3) to generalize—i.e., guessing where probability function mass/density concentrates, (4) to fight the curse of dimensionality, and (5) to disentangle underlying explanatory factors of the data—i.e., to make sense of the data in the context of an application domain. To address these challenges and to ensure successful machine learning applications in various domains an integrated machine learning approach is important. This requires a concerted international effort without boundaries, supporting collaborative, cross-domain, interdisciplinary and transdisciplinary work of experts from seven sections, ranging from data pre-processing to data visualization, i.e., to map results found in arbitrarily high dimensional spaces into the lower dimensions to make it accessible, usable and useful to the end user. An integrated machine learning approach needs also to consider issues of privacy, data protection, safety, security, user acceptance and social implications. This paper is the inaugural introduction to the new journal of MAchine Learning & Knowledge Extraction (MAKE). The goal is to provide an incomplete, personally biased, but consistent introduction into the concepts of MAKE and a brief overview of some selected topics to stimulate future research in the international research community.

Identification of Long Bone Fractures in Radiology Reports Using Natural Language Processing to support Healthcare Quality Improvement

BACKGROUND

Important information to support healthcare quality improvement is often recorded in free text documents such as radiology reports. Natural language processing (NLP) methods may help extract this information, but these methods have rarely been applied outside the research laboratories where they were developed.

OBJECTIVE

To implement and validate NLP tools to identify long bone fractures for pediatric emergency medicine quality improvement.

METHODS

Using freely available statistical software packages, we implemented NLP methods to identify long bone fractures from radiology reports. A sample of 1,000 radiology reports was used to construct three candidate classification models. A test set of 500 reports was used to validate the model performance. Blinded manual review of radiology reports by two independent physicians provided the reference standard. Each radiology report was segmented and word stem and bigram features were constructed. Common English "stop words" and rare features were excluded. We used 10-fold cross-validation to select optimal configuration parameters for each model. Accuracy, recall, precision and the F1 score were calculated. The final model was compared to the use of diagnosis codes for the identification of patients with long bone fractures.

RESULTS

There were 329 unique word stems and 344 bigrams in the training documents. A support vector machine classifier with Gaussian kernel performed best on the test set with accuracy=0.958, recall=0.969, precision=0.940, and F1 score=0.954. Optimal parameters for this model were cost=4 and gamma=0.005. The three classification models that we tested all performed better than diagnosis codes in terms of accuracy, precision, and F1 score (diagnosis code accuracy=0.932, recall=0.960, precision=0.896, and F1 score=0.927).

CONCLUSIONS

NLP methods using a corpus of 1,000 training documents accurately identified acute long bone fractures from radiology reports. Strategic use of straightforward NLP methods, implemented with freely available software, offers quality improvement teams new opportunities to extract information from narrative documents.