Impact of different visualisation of AI output

explainability_paper_

There are different approaches taken by AI vendors to visualise AI output to the end user. An interesting aspect of these visualisations is that the way in which AI results are presented, and the level of explainability embedded, can have a significant impact on radiologists' diagnostic performance and confidence in the technology. A recent study by Prinster et al (2024) explored this phenomenon in the context of chest radiography diagnosis, comparing the impact of two types of explanation: local (feature-based) and global (prototype-based).

The study aimed to assess how explanation type, AI confidence, and AI output correctness influenced diagnostic accuracy, efficiency, and trust. Participants were tasked with interpreting chest radiographs assisted by simulated AI suggestions.

They conducted a multicenter, randomized study involving 220 physicians, who were presented with AI-assisted diagnostic tasks using local and global explanations. The research showed that local explanations improved diagnostic accuracy when AI advice was correct and increased efficiency by reducing time spent on interpretation. However, they also led to greater “simple trust,” or reliance on AI advice, regardless of its correctness, which could risk overreliance on incorrect recommendations. Read the full study for a deep dive into the impact of visualization and the explainability of AI output.

Read full study


Care to Explain? AI Explanation Types Differentially Impact Chest Radiograph Diagnostic Performance and Physician Trust in AI

Radiology, 2024

Abstract

Background

It is unclear whether artificial intelligence (AI) explanations help or hurt radiologists and other physicians in AI-assisted radiologic diagnostic decision-making.

Purpose

To test whether the type of AI explanation and the correctness and confidence level of AI advice impact physician diagnostic performance, perception of AI advice usefulness, and trust in AI advice for chest radiograph diagnosis.

Materials and Methods

A multicenter, prospective randomized study was conducted from April 2022 to September 2022. Two types of AI explanations prevalent in medical imaging—local (feature-based) explanations and global (prototype-based) explanations—were a between-participant factor, while AI correctness and confidence were within-participant factors. Radiologists (task experts) and internal or emergency medicine physicians (task nonexperts) received a chest radiograph to read; then, simulated AI advice was presented. Generalized linear mixed-effects models were used to analyze the effects of the experimental variables on diagnostic accuracy, efficiency, physician perception of AI usefulness, and “simple trust” (ie, speed of alignment with or divergence from AI advice); the control variables included knowledge of AI, demographic characteristics, and task expertise. Holm-Sidak corrections were used to adjust for multiple comparisons.

Results

Data from 220 physicians (median age, 30 years [IQR, 28–32.75 years]; 146 male participants) were analyzed. Compared with global AI explanations, local AI explanations yielded better physician diagnostic accuracy when the AI advice was correct (β = 0.86; P value adjusted for multiple comparisons [Padj] < .001) and increased diagnostic efficiency overall by reducing the time spent considering AI advice (β = −0.19; Padj = .01). While there were interaction effects of explanation type, AI confidence level, and physician task expertise on diagnostic accuracy (β = −1.05; Padj = .04), there was no evidence that AI explanation type or AI confidence level significantly affected subjective measures (physician diagnostic confidence and perception of AI usefulness). Finally, radiologists and nonradiologists placed greater simple trust in local AI explanations than in global explanations, regardless of the correctness of the AI advice (β = 1.32; Padj = .048).

Conclusion

The type of AI explanation impacted physician diagnostic performance and trust in AI, even when physicians themselves were not aware of such effects.