Machine Learning for Health Informatics

Abstract

Abstract. Machine Learning (ML) studies algorithms which can learn from data to gain knowledge from experience and to make decisions and predictions. Health Informatics (HI) studies the effective use of probabilistic information for decision making. The combination of both has greatest potential to rise quality, efficacy and efficiency of treatment and care. Health systems worldwide are confronted with "big data" in high dimensions, where the inclusion of a human is impossible and automatic ML (aML) show impressive results. However, sometimes we are confronted with complex data, "little data", or rare events, where aMLapproaches suffer of insufficient training samples. Here interactive ML (iML) may be of help, particularly with a doctor-in-the-loop, e.g. in subspace clustering, k-Anonymization, protein folding and protein design. However, successful application of ML for HI needs an integrated approach, fostering a concerted effort of four areas: (1) data science, (2) algorithms (with focus on networks and topology (structure), and entropy (time), (3) data visualization, and last but not least (4) privacy, data protection, safety & security.

Methods for Segmentation and Classification of Digital Microscopy Tissue Images

“…Human involvement is still a critical component in effective application of deep learning and other automated techniques in biomedical image analysis. Human expert review of image analysis results is important not only for quality assessment and interpretation of analysis results but also for iterative or active improvement of segmentation and classification models (and producing increasingly more accurate and robust results) (Holzinger, 2016; Holzinger et al, 2018). One of the challenges to human-in-the-loop in pathology image analysis is the sheer volume of data.…”

Breathing Rate Estimation from Head-Worn Photoplethysmography Sensor Data Using Machine Learning

“…Even though ML has already infiltrated many domains of health informatics [ 26 ], its efficiency in the field of breathing rate estimation from wearable sensors data has not been thoroughly explored; studies that have attempted to estimate breathing rate from PPG data using ML have been scarcely published. Shuzan et al [ 27 ] proposed an ML method for breathing rate estimation from PPG data based on Gaussian process regression.…”

Human–Machine Interaction in Driving Assistant Systems for Semi-Autonomous Driving Vehicles

“…Mantini et al extracted bio-signal features using power spectrum density (PSD) [48], and Topic et al used topography [49]. However, it is difficult to apply real-time emotion recognition because of the time delay required in the feature extraction process [50,51]. We used PPG [52] and GSR [53] bio-signals with specific regularity, which are easy to acquire in real time.…”