Mutual Explanations for Cooperative Decision Making in Medicine

Schmid, Ute; Finzel, Bettina

doi:10.1007/s13218-020-00633-2

Cited by 32 publications

(57 citation statements)

References 24 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For example, if tumor tissue has grown passed muscle tissue and already invades fat, the tumor class is more critical compared to a tumor that resides within tissue of the mucosa (Wittekind, 2016). As further pointed out in Schmid and Finzel (2020), ML approaches should therefore be able to reveal which relationships lead to a certain classification. Furthermore, relationships should be communicated in a comprehensible way to medical experts and this can be achieved with the help of natural language explanations.…”

Section: Verbal Explanationsmentioning

confidence: 97%

“…In addition, experts can still provide their knowledge to the algorithm, as illustrated in Figure 8, where an exemplary spatial relationship touches is defined in the background knowledge FIGURE 7 | Training examples and learned rules for a hypothetical diagnostic domain of colon cancer (Schmid and Finzel, 2020).…”

Section: Verbal Explanationsmentioning

confidence: 99%

“…FIGURE 8 | Background theory with domain rules for a hypothetical diagnostic domain of colon cancer (Schmid and Finzel, 2020). and can thus be found by the algorithm in the data if relevant to the classification of pT3.…”

Section: Verbal Explanationsmentioning

confidence: 99%

“…Adaptive machine learning, which incorporates interaction with the human and incremental learning, is suitable to enhance such companions (Siebers and Schmid, 2019). Furthermore, cognitive conditions imposed by the context FIGURE 9 | Model of a mutual explanations system (Schmid and Finzel, 2020).…”

Section: Interactive Machine Learningmentioning

confidence: 99%

“…As described in the previous subsection, providing visual explanations and semantics helps to increase the interpretability of opaque classifiers. In addition, natural language explanations constitute an important explanation modality, since, for their expressiveness, they capture complex relationships better than visualizations (Finzel et al, 2019 ; Rabold et al, 2019 ; Schmid and Finzel, 2020 ) and increase comprehensibility (Muggleton et al, 2018 ).…”

Section: Fundamentals Of Cai Transition Applied To Medicinementioning

confidence: 99%

See 4 more Smart Citations

The Next Generation of Medical Decision Support: A Roadmap Toward Transparent Expert Companions

Bruckert

Finzel

Schmid

2020

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

Increasing quality and performance of artificial intelligence (AI) in general and machine learning (ML) in particular is followed by a wider use of these approaches in everyday life. As part of this development, ML classifiers have also gained more importance for diagnosing diseases within biomedical engineering and medical sciences. However, many of those ubiquitous high-performing ML algorithms reveal a black-box-nature, leading to opaque and incomprehensible systems that complicate human interpretations of single predictions or the whole prediction process. This puts up a serious challenge on human decision makers to develop trust, which is much needed in life-changing decision tasks. This paper is designed to answer the question how expert companion systems for decision support can be designed to be interpretable and therefore transparent and comprehensible for humans. On the other hand, an approach for interactive ML as well as human-in-the-loop-learning is demonstrated in order to integrate human expert knowledge into ML models so that humans and machines act as companions within a critical decision task. We especially address the problem of Semantic Alignment between ML classifiers and its human users as a prerequisite for semantically relevant and useful explanations as well as interactions. Our roadmap paper presents and discusses an interdisciplinary yet integrated Comprehensible Artificial Intelligence (cAI)-transition-framework with regard to the task of medical diagnosis. We explain and integrate relevant concepts and research areas to provide the reader with a hands-on-cookbook for achieving the transition from opaque black-box models to interactive, transparent, comprehensible and trustworthy systems. To make our approach tangible, we present suitable state of the art methods with regard to the medical domain and include a realization concept of our framework. The emphasis is on the concept of Mutual Explanations (ME) that we introduce as a dialog-based, incremental process in order to provide human ML users with trust, but also with stronger participation within the learning process.

show abstract

Section: Verbal Explanationsmentioning

confidence: 97%

Section: Verbal Explanationsmentioning

confidence: 99%

Section: Verbal Explanationsmentioning

confidence: 99%

Section: Interactive Machine Learningmentioning

confidence: 99%

Section: Fundamentals Of Cai Transition Applied To Medicinementioning

confidence: 99%

See 3 more Smart Citations

The Next Generation of Medical Decision Support: A Roadmap Toward Transparent Expert Companions

Bruckert

Finzel

Schmid

2020

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

show abstract

Expressive Explanations of DNNs by Combining Concept Analysis with ILP

Rabold

Schwalbe

Schmid

2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Explainable AI has emerged to be a key component for blackbox machine learning approaches in domains with a high demand for reliability or transparency. Examples are medical assistant systems, and applications concerned with the General Data Protection Regulation of the European Union, which features transparency as a cornerstone. Such demands require the ability to audit the rationale behind a classifier's decision. While visualizations are the de facto standard of explanations, they come short in terms of expressiveness in many ways: They cannot distinguish between different attribute manifestations of visual features (e.g. eye open vs. closed), and they cannot accurately describe the influence of absence of, and relations between features. An alternative would be more expressive symbolic surrogate models. However, these require symbolic inputs, which are not readily available in most computer vision tasks. In this paper we investigate how to overcome this: We use inherent features learned by the network to build a global, expressive, verbal explanation of the rationale of a feed-forward convolutional deep neural network (DNN). The semantics of the features are mined by a concept analysis approach trained on a set of human understandable visual concepts. The explanation is found by an Inductive Logic Programming (ILP) method and presented as first-order rules. We show that our explanation is faithful to the original black-box model 3 .

show abstract

Progressive Intensity of Human-Technology Teaming

Waefler

2021

Human Interaction, Emerging Technologies and Future Systems V

View full text Add to dashboard Cite

Mutual Explanations for Cooperative Decision Making in Medicine

Cited by 32 publications

References 24 publications

The Next Generation of Medical Decision Support: A Roadmap Toward Transparent Expert Companions

The Next Generation of Medical Decision Support: A Roadmap Toward Transparent Expert Companions

Expressive Explanations of DNNs by Combining Concept Analysis with ILP

Progressive Intensity of Human-Technology Teaming

Contact Info

Product

Resources

About