AI clinical

Podcast: Clinicians’ ‘Number-One Wish’ for Artificial Intelligence

…we installed cheap depth sensors that can collect human behavior data on patients and clinicians without infringing on their privacy, because these are not photo grabs of people’s faces and identities. With that information, we can observe longitudinally, 24/7, whether proper care is being given to our patients and provide feedback in the health delivery system.

Topol, E. & Li, F. (2020). Clinicians’ ‘Number-One Wish’ for Artificial Intelligence. Medicine and the Machine podcast. Medscape.

I didn’t hear what the number one wish was (I was driving to work and may have been distracted for a moment) but the conversation is generally worth listening to. Topol and Li both have good insight into the application of AI in clinical contexts and the conversation touches on some of the technical aspects of AI (e.g. bias, training machine learning algorithms, labeled datasets, etc.) while staying accessible for listeners who are unfamiliar with the details.

One of the standout bits for me was the discussion around how the use of depth sensors in an ICU can generate data that an AI can use to map the behaviour of staff within the unit, to the extent that it can tell whether or not basic levels of care are being met. You might have concerns about issues of privacy and the surveillance of staff but if one of my family members were in an ICU, I know that I’d want to know if everyone is washing their hands appropriately.

The link above includes a transcript of the conversation.

AI clinical

AI outperforms clinicians in triaging post-operative patients for ICUe.

Artificial intelligence correctly triaged 41 of the 50 patients in the study (82%). Surgeons had an accuracy triage rate of 70% (35 patients), intensivists 64% (32 patients), and anaesthesiologists 58% (29 patients). The number of incorrect triage decisions was lowest for AI (18%), followed by 30% for surgeons, 36% for intensivists, and 42% for anaesthesiologists.

Editor’s pick, (2019). AI outperforms clinicians in triaging post-operative patients for ICUe. Medical brief.

These are the kinds of contexts where we’ll increasingly see the use of machine learning algorithms to “provide guidance” to clinicians: high stakes decision-making scenarios where the correct outcome relies on the integration of data from a wide variety of clinical domains that are not optimised for human cognition. It’s just not possible for a human being – or team of human beings – to track the high number of relevant and inter-related variables that influence these kinds of clinical outcomes.

The resulting algorithm included 87 clinical variables and 15 specific criteria related to admission to the ICU within 48 hours of surgery.


Proposal abstract: Training in the ICU for physiotherapy students with a visual impairment (a case study)

Abstract for a project proposal that I submitted for ethics review earlier this week. If it gets approved we’ll begin data collection on our first visually impaired undergraduate student placement in the intensive care unit.

The Department of Physiotherapy at the University of the Western Cape (UWC) began accepting students with visual impairments (VI) into the undergraduate programme in 1996. To date, eight students with visual impairments have graduated with degrees in physiotherapy, all of whom have gone on to successful employment in the health system. In this area, the department has played an important role in leading transformative change, not only in the broader context of higher education but specifically in the area of providing equal opportunities for professional training for all South Africans.

While the department has done well to provide equal opportunities to students with VI in the general undergraduate course, we have yet to place a student with VI into the ICU setting as part of their clinical rotations. Early in 2015 however, the department engaged in a series of discussions with one of our final year students with VI, as well as clinicians and lecturers and decided to explore the possibility of placing the student into the ICU. This would enable us to align ourselves with national policies and priorities. As part of this process of placing a student in the ICU setting we want to describe the facilitators and barriers that exist, as seen from the perspective of those involved in the process. The aim of the study is therefore to explore the experiences of the student, clinicians, academics and peers, with the placement of a student with a VI in the ICU.

How do you negotiate this environment when you can’t see very well?

The project will make use of a case study design that aims to describe the process of placing an undergraduate physiotherapy student with VI in an ICU setting as part of a clinical practice rotation. The case study will include data gathered from the student’s reflective clinical diary as well as in-depth interviews with the student, clinical supervisor, VI and clinical coordinators in the physiotherapy department at UWC, and the clinician who is responsible for overseeing the student in the ICU. Peers who have engaged with the student during the specific clinical placement will also be included, and will be identified during the process.

The interviews will be audio recorded and then sent away for transcription. The transcribed interviews will be anonymised and thematically analysed in order to determine themes related to barriers and facilitators that are relevant to the student’s learning. The transcripts – along with the analyses – will be shared with participants in order to ensure that the themes that emerged are consistent with the meaning that they had intended during the interviews.