Search

Medical use cases for machine learning (ML) are growing exponentially. The first hospitals are already using ML systems as decision support systems in their daily routine. At the same time, most ML systems are still opaque and it is not clear how these systems arrive at their predictions. In this paper, we provide a brief overview of the taxonomy of explainability methods and review popular methods. In addition, we conduct a systematic literature search on PubMed to investigate which explainable artificial intelligence (XAI) methods are used in 450 specific medical supervised ML use cases, how the use of XAI methods has emerged recently, and how the precision of describing ML pipelines has evolved over the past 20 years. A large fraction of publications with ML use cases do not use XAI methods at all to explain ML predictions. However, when XAI methods are used, open-source and model-agnostic explanation methods are more commonly used, with SHapley Additive exPlanations (SHAP) and Gradient Class Activation Mapping (Grad-CAM) for tabular and image data leading the way. ML pipelines have been described in increasing detail and uniformity in recent years. However, the willingness to share data and code has stagnated at about one-quarter. XAI methods are mainly used when their application requires little effort. The homogenization of reports in ML use cases facilitates the comparability of work and should be advanced in the coming years. Experts who can mediate between the worlds of informatics and medicine will become more and more in demand when using ML systems due to the high complexity of the domain. [Display omitted] • We estimate that only 16 % of the reported explainability methods could be understood by patients. • The distribution of data types for explainable ML applications for tabular | image | text | audio data is 51 % | 32 % | 3 % | 0 %. • The quality of the description of machine learning pipelines increased in recent years with more homogeneity. • The data and code sharing ratio stagnated in about one quarter. • Most popular explainability methods are SHAP, LIME, and Grad-CAM. [ABSTRACT FROM AUTHOR]

Background: Over the past 30 years, the international conference on Artificial Intelligence in MEdicine (AIME) has been organized at different venues across Europe every 2 years, establishing a forum for scientific exchange and creating an active research community. The Artificial Intelligence in Medicine journal has published theme issues with extended versions of selected AIME papers since 1998.Objectives: To review the history of AIME conferences, investigate its impact on the wider research field, and identify challenges for its future.Methods: We analyzed a total of 122 session titles to create a taxonomy of research themes and topics. We classified all 734 AIME conference papers published between 1985 and 2013 with this taxonomy. We also analyzed the citations to these conference papers and to 55 special issue papers.Results: We identified 30 research topics across 12 themes. AIME was dominated by knowledge engineering research in its first decade, while machine learning and data mining prevailed thereafter. Together these two themes have contributed about 51% of all papers. There have been eight AIME papers that were cited at least 10 times per year since their publication.Conclusions: There has been a major shift from knowledge-based to data-driven methods while the interest for other research themes such as uncertainty management, image and signal processing, and natural language processing has been stable since the early 1990s. AIME papers relating to guidelines and protocols are among the most highly cited. [ABSTRACT FROM AUTHOR]

Summary: Objective: Medical data is often very high dimensional. Depending upon the use, some data dimensions might be more relevant than others. In processing medical data, choosing the optimal subset of features is such important, not only to reduce the processing cost but also to improve the usefulness of the model built from the selected data. This paper presents a data mining study of medical data with fuzzy modeling methods that use feature subsets selected by some indices/methods. Methods: Specifically, three fuzzy modeling methods including the fuzzy k-nearest neighbor algorithm, a fuzzy clustering-based modeling, and the adaptive network-based fuzzy inference system are employed. For feature selection, a total of 11 indices/methods are used. Medical data mined include the Wisconsin breast cancer dataset and the Pima Indians diabetes dataset. The classification accuracy and computational time are reported. To show how good the best performer is, the globally optimal was also found by carrying out an exhaustive testing of all possible combinations of feature subsets with three features. Results: For the Wisconsin breast cancer dataset, the best accuracy of 97.17% was obtained, which is only 0.25% lower than that was obtained by exhaustive testing. For the Pima Indians diabetes dataset, the best accuracy of 77.65% was obtained, which is only 0.13% lower than that obtained by exhaustive testing. Conclusion: This paper has shown that feature selection is important to mining medical data for reducing processing time and for increasing classification accuracy. However, not all combinations of feature selection and modeling methods are equally effective and the best combination is often data-dependent, as supported by the breast cancer and diabetes data analyzed in this paper. [Copyright &y& Elsevier]

This paper describes the fuzzy knowledge representation framework of the medical computer consultation system MedFrame/CADIAG-IV as well as the specific knowledge acquisition techniques that have been developed to support the definition of knowledge concepts and inference rules. As in its predecessor system CADIAG-II, fuzzy medical knowledge bases are used to model the uncertainty and the vagueness of medical concepts and fuzzy logic reasoning mechanisms provide the basic inference processes. The elicitation and acquisition of medical knowledge from domain experts has often been described as the most difficult and time-consuming task in knowledge-based system development in medicine. It comes as no surprise that this is even more so when unfamiliar representations like fuzzy membership functions are to be acquired. From previous projects we have learned that a user-centered approach is mandatory in complex and ill-defined knowledge domains such as internal medicine. This paper describes the knowledge acquisition framework that has been developed in order to make easier and more accessible the three main tasks of: (a) defining medical concepts; (b) providing appropriate interpretations for patient data; and (c) constructing inferential knowledge in a fuzzy knowledge representation framework. Special emphasis is laid on the motivations for some system design and data modeling decisions. The theoretical framework has been implemented in a software package, the Knowledge Base Builder Toolkit. The conception and the design of this system reflect the need for a user-centered, intuitive, and easy-to-handle tool. First results gained from pilot studies have shown that our approach can be successfully implemented in the context of a complex fuzzy theoretical framework. As a result, this critical aspect of knowledge-based system development can be accomplished more easily. [Copyright &y& Elsevier]

Artificial intelligence is a broad field that comprises a wide range of techniques, where deep learning is presently the one with the most impact. Moreover, the medical field is an area where data both complex and massive and the importance of the decisions made by doctors make it one of the fields in which deep learning techniques can have the greatest impact. A systematic review following the Cochrane recommendations with a multidisciplinary team comprised of physicians, research methodologists and computer scientists has been conducted. This survey aims to identify the main therapeutic areas and the deep learning models used for diagnosis and treatment tasks. The most relevant databases included were MedLine, Embase, Cochrane Central, Astrophysics Data System, Europe PubMed Central, Web of Science and Science Direct. An inclusion and exclusion criteria were defined and applied in the first and second peer review screening. A set of quality criteria was developed to select the papers obtained after the second screening. Finally, 126 studies from the initial 3493 papers were selected and 64 were described. Results show that the number of publications on deep learning in medicine is increasing every year. Also, convolutional neural networks are the most widely used models and the most developed area is oncology where they are used mainly for image analysis. [ABSTRACT FROM AUTHOR]

Summary: Objective: This paper proposes a novel method for the extraction and classification of individual motor unit action potentials (MUAPs) from intramuscular electromyographic signals. Methodology: The proposed method automatically detects the number of template MUAP clusters and classifies them into normal, neuropathic or myopathic. It consists of three steps: (i) preprocessing of electromyogram (EMG) recordings, (ii) MUAP detection and clustering and (iii) MUAP classification. Results: The approach has been validated using a dataset of EMG recordings and an annotated collection of MUAPs. The correct identification rate for MUAP clustering is 93, 95 and 92% for normal, myopathic and neuropathic, respectively. Ninety-one percent of the superimposed MUAPs were correctly identified. The obtained accuracy for MUAP classification is about 86%. Conclusion: The proposed method, apart from efficient EMG decomposition addresses automatic MUAP classification to neuropathic, myopathic or normal classes directly from raw EMG signals. [Copyright &y& Elsevier]

In this paper, the current published knowledge about smart and adaptive engineering systems in medicine is reviewed. The achievements of frontier research in this particular field within medical engineering are described. A multi-disciplinary approach to the applications of adaptive systems is observed from the literature surveyed. The three modalities of diagnosis, imaging and therapy are considered to be an appropriate classification method for the analysis of smart systems being applied to specified medical sub-disciplines. It is expected that future research in biomedicine should identify subject areas where more advanced intelligent systems could be applied than is currently evident. The literature provides evidence of hybridisation of different types of adaptive and smart systems with applications in different areas of medical specifications. [Copyright &y& Elsevier]