Limits of trust in medical AI

2020 ◽  
Vol 46 (7) ◽  
pp. 478-481 ◽  
Author(s):  
Joshua James Hatherley
Keyword(s):  
Artificial Intelligence ◽  
Decision Making ◽  
Diabetic Retinopathy ◽  
Deep Learning ◽  
Clinical Practice ◽  
Medical Decision Making ◽  
Hospital Readmissions ◽  
New Drugs ◽  
Medical Decision ◽  
Practice Of Medicine

Artificial intelligence (AI) is expected to revolutionise the practice of medicine. Recent advancements in the field of deep learning have demonstrated success in variety of clinical tasks: detecting diabetic retinopathy from images, predicting hospital readmissions, aiding in the discovery of new drugs, etc. AI’s progress in medicine, however, has led to concerns regarding the potential effects of this technology on relationships of trust in clinical practice. In this paper, I will argue that there is merit to these concerns, since AI systems can be relied on, and are capable of reliability, but cannot be trusted, and are not capable of trustworthiness. Insofar as patients are required to rely on AI systems for their medical decision-making, there is potential for this to produce a deficit of trust in relationships in clinical practice.

May We Support Your Decision?

1996 ◽  
Vol 1 (3) ◽  
pp. 175-178 ◽  
Author(s):  
Colin Gordon
Keyword(s):  
Artificial Intelligence ◽  
Health Care ◽  
Decision Making ◽  
Clinical Practice ◽  
Decision Support ◽  
Health Care Professionals ◽  
Medical Decision Making ◽  
Electronic Patient Record ◽  
Medical Decision ◽  
Technical Developments

Expert systems to support medical decision-making have so far achieved few successes. Current technical developments, however, may overcome some of the limitations. Although there are several theoretical currents in medical artificial intelligence, there are signs of them converging. Meanwhile, decision support systems, which set themselves more modest goals than replicating or improving on clinicians' expertise, have come into routine use in places where an adequate electronic patient record exists. They may also be finding a wider role, assisting in the implementation of clinical practice guidelines. There is, however, still much uncertainty about the kinds of decision support that doctors and other health care professionals are likely to want or accept.

The Contextual Nature of Critical Care Judgment

1994 ◽  
Vol 9 (2) ◽  
pp. 58-63 ◽  
Author(s):  
Gilbert M. Goldman ◽  
Thyyar M. Ravindranath
Keyword(s):  
Decision Making ◽  
Critical Care ◽  
Clinical Practice ◽  
Clinical Judgment ◽  
Retrospective Review ◽  
Medical Decision Making ◽  
Formal Analysis ◽  
Medical Decision ◽  
Time Constraints ◽  
Care Decision

Critical care decision-making involves principles common to all medical decision-making. However, critical care is a remarkably distinctive form of clinical practice and therefore it may be useful to distinguish those elements particularly important or unique to ICU decision-making. The peculiar contextuality of critical care decision-making may be the best example of these elements. If so, attempts to improve our understanding of ICU decision-making may benefit from a formal analysis of its remarkable contextual nature. Four key elements of the context of critical care decisions can be identified: (1) costs, (2) time constraints, (3) the uncertain status of much clinical data, and (4) the continually changing environment of the ICU setting. These 4 elements comprise the context for the practice of clinical judgment in the ICU. The fact that intensivists are severely constrained by teh context of each case has important ramifications both for practice and for retrospective review. During retrospective review, the contextual nature of ICU judgment may be unfairly neglected by ignoring one or more of the key elements. Such neglect can be avoided if intensivists demand empathetic evaluation from reviewers.

scholarly journals Technology of Supporting Medical Decision-Making Using Evidence-Based Medicine and Artificial Intelligence

2020 ◽  
Vol 176 ◽  
pp. 1703-1712
Author(s):  
Georgy Lebedev ◽  
Eduard Fartushnyi ◽  
Igor Fartushnyi ◽  
Igor Shaderkin ◽  
Herman Klimenko ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Decision Making ◽  
Evidence Based Medicine ◽  
Medical Decision Making ◽  
Medical Decision ◽  
Evidence Based ◽  
Based Medicine

Assessing medical decision making capacity among cancer patients: Preliminary clinical experience of using a competency assessment instrument

2013 ◽  
Vol 13 (6) ◽  
pp. 1529-1533 ◽  
Author(s):  
Tatsuo Akechi ◽  
Toru Okuyama ◽  
Megumi Uchida ◽  
Koji Sugano ◽  
Yosuke Kubota ◽  
...  
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Cancer Patients ◽  
Medical Decision Making ◽  
Past History ◽  
Structured Interview ◽  
Assessment Instrument ◽  
Medical Decision ◽  
Dementia Patient ◽  
Decision Making Capacity

AbstractObject:This study investigates the usefulness of the Structured Interview for Competency and Incompetency Assessment Testing and Ranking Inventory (SICIATRI) for cancer patients, which is a structured interview that assesses a patient's competency in clinical practice.Methods:The SICIATRI, originally developed to measure patients' competency to give informed consent, were administered referred cancer patients who needed for assessing medical decision making capacity. The usefulness of the SICIATRI was investigated retrospectively. Recommendation for modification of the SICIATRI for cancer patients if applicable were made by the research team.Results:Among the 433 cancer patients referred for psychiatric consultation, 12 were administered the SICIATRI and all of the administration were conducted without big problems. All patients were 60 years or older. The most common purpose for competency evaluation was to analyze patients' understanding of the anti-cancer treatment proposed by oncologists, followed by their refusal of the treatment. Half of the patients (n = 6) were diagnosed with delirium and three among them were judged as having the most impaired status of a patient's competency. Two patients (17%) were diagnosed with major depression and another two (17%) were mental retardation and each one patient was diagnosed with dementia and past history of alcohol dependence. Among 6 patients without delirium 5 subjects including a dementia patient were judged as fully competent. Total of 5 small potential modifications of the SICIATRI for its use with Japanese cancer patients were recommended.Significance of results:Our experience suggests that the SICIATRI is a useful instrument for psycho-oncology clinical practice.

scholarly journals Counterpoint To (measure apo)B or not to (measure apo)B: a critique of modern medical decision-making

1997 ◽  
Vol 43 (8) ◽  
pp. 1310-1314 ◽  
Author(s):  
Allan D Sniderman
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Scientific Method ◽  
Medical Decision Making ◽  
Pharmacological Therapy ◽  
Medical Decision ◽  
Apo B ◽  
New Approach ◽  
Consensus Conferences ◽  
Lipoprotein Lipids

Abstract The measurement of apo B provides critical information that is complementary to that provided by the plasma and lipoprotein lipids for the assessment of coronary risk and the choice of appropriate pharmacological therapy. Why then is this measurement not in more widespread clinical use? I suggest two explanations. First, against the evidence, there is a lingering perception that problems persist in its measurement in routine clinical practice. Far from this being the case, however, the measurement of apo B has met every reasonable standard of laboratory precision and reliability to allow its widespread introduction in clinical laboratories. The second impediment is that the introduction of new tests has become subject to the authority of consensus conferences, a new approach to medical decision-making. The number of such conferences is increasing astronomically, and their reports are major determinants of clinical practice and allocation of resources. Notwithstanding the benefits they have brought, here I argue that, just as with any other scientific method, the merits of this new method of decision-making need to be examined critically; for if we do not, a process that was established to introduce change may, in fact, retard it or destroy it altogether.

Modeling Deep Learning Neural Networks With Denotational Mathematics in UbiHealth Environment

2020 ◽  
Vol 12 (3) ◽  
pp. 14-27 ◽  
Author(s):  
John Sarivougioukas ◽  
Aristides Vagelatos
Keyword(s):  
Decision Making ◽  
Deep Learning ◽  
Ubiquitous Computing ◽  
Medical Decision Making ◽  
Medical Decision ◽  
Learning Networks ◽  
Healthcare Applications ◽  
Medical Diagnoses ◽  
The Given

Ubiquitous computing environments that are involved in healthcare applications are typically characterized by dynamically changing contexts. The contextual information must be efficiently processed in order to support medical decision making. The ubiquitous computing healthcare ecosystem must be capable of extracting medically valuable characteristics, making precise decisions, and taking medically appropriate actions. In this framework, deep learning networks can be used for data fusion of large and complex sets of information in order to make the appropriate medical diagnoses. The quality of decisions depends on the selection of appropriate network weights, which define a transformation of the given input into a diagnosis. Denotational mathematics provide a promising framework for modeling deep learning networks and adjusting their behavior by adapting their weights for the given input. Furthermore, the fidelity of the network's output can be controlled by applying a regulator to the weights values. The authors show that Denotational Mathematics can serve as a rigorous framework for modeling and controlling deep learning networks, thereby enhancing the quality of medical decision making.

scholarly journals Artificial Intelligence in Cardiac Imaging

2020 ◽  
Vol 13 (2) ◽  
pp. 110-116 ◽  
Author(s):  
Karthik Seetharam ◽  
Sirish Shrestha ◽  
Partho P Sengupta
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Decision Making ◽  
Cardiac Imaging ◽  
Medical Decision Making ◽  
Medical Decision ◽  
Precision Phenotyping ◽  
Cardiac Disorders ◽  
Enhance Patient Care

Machine learning (ML), a subset of artificial intelligence, is showing promising results in cardiology, especially in cardiac imaging. ML algorithms are allowing cardiologists to explore new opportunities and make discoveries not seen with conventional approaches. This offers new opportunities to enhance patient care and open new gateways in medical decision-making. This review highlights the role of ML in cardiac imaging for precision phenotyping and prognostication of cardiac disorders.

scholarly journals Liability issues for the use of artificial intelligence in health care in Canada: AI and medical decision-making

2020 ◽  
Vol 46 (2) ◽  
Author(s):  
Mélanie Bourassa Forcier ◽  
Lara Khoury ◽  
Nathalie Vézina
Keyword(s):  
Artificial Intelligence ◽  
Health Care ◽  
Decision Making ◽  
Medical Decision Making ◽  
Legal Framework ◽  
Medical Decision

This paper explores Canadian liability concerns flowing from the integration of artificial intelligence (AI) as a tool assisting physicians in their medical decision-making. It argues that the current Canadian legal framework is sufficient, in most cases, to allow developers and users of AI technology to assess each stakeholder's responsibility should the technology cause harm.

Sign in / Sign up

Export Citation Format

Share Document