Computerized Decision Support for Bladder Cancer Treatment Response Assessment in CT Urography: Effect on Diagnostic Accuracy in Multi-Institution Multi-Specialty Study.

Di Sun,Yousef Zakharia,Lauren Pomerantz,Richard H Cohan,Sean Woolen,Dean Elhag,Ajjai Alva,Ravi K Samala,Alon Z Weizer,Chuan Zhou,Wesley T Kerr,Isaac R Francis,Rohan Garje,Matthew S Davenport,Daniel Barkmeier,Monika Joshi,Elaine M Caoili,Galina Kirova-Nedyalkova,Phillip L Palmbos,Heang-Ping Chan,Prasad R Shankar,Nathaniel Meyer,H Kenny ,Lubomir M Hadjiiski ,Kimberly L Shampain ,Martha M Matuszak

doi:10.3390/tomography8020054

Abstract

This observer study investigates the effect of computerized artificial intelligence (AI)-based decision support system (CDSS-T) on physicians’ diagnostic accuracy in assessing bladder cancer treatment response. The performance of 17 observers was evaluated when assessing bladder cancer treatment response without and with CDSS-T using pre- and post-chemotherapy CTU scans in 123 patients having 157 pre- and post-treatment cancer pairs. The impact of cancer case difficulty, observers’ clinical experience, institution affiliation, specialty, and the assessment times on the observers’ diagnostic performance with and without using CDSS-T were analyzed. It was found that the average performance of the 17 observers was significantly improved (p = 0.002) when aided by the CDSS-T. The cancer case difficulty, institution affiliation, specialty, and the assessment times influenced the observers’ performance without CDSS-T. The AI-based decision support system has the potential to improve the diagnostic accuracy in assessing bladder cancer treatment response and result in more consistent performance among all physicians.

Highlights

In 2021, it was estimated that about 83,730 new cases of bladder cancer would be diagnosed and about 17,200 would die from it in the US
For the overall performance evaluation (17 observers and 157 cancers), the results showed that the CDSS-T aid can improve the observers’ performance significantly in assessing bladder treatment response
While the other groups had significant improvement for easy subset but not for difficult subset, the five oncologists only had significant improvement for difficult subset, from area under curve (AUC) of 0.57 without CDSS-T to 0.63 with CDSS-T. This happened because only oncologists reached AUC of 0.63 with CDSS-T which was the best, while the other two groups were 0.60~0.62, and it increased from AUC of 0.57 which was the lowest

Summary

Introduction

In 2021, it was estimated that about 83,730 new cases of bladder cancer would be diagnosed and about 17,200 would die from it in the US. This would account for 0.4% of all new cancer cases and 2.8% of all cancer deaths [1]. The 5-year relative survival rate for all SEER stages combined is 77% [1]. Neoadjuvant chemotherapy performed prior to radical cystectomy can improve patient survival rate and decrease the probability of metastatic disease [2–4]. It is of great importance to evaluate the response of bladder lesions to chemotherapy treatment to spare the patient the toxicities of further unnecessary chemotherapy or to support surgery de-escalation [6]

Objectives

Methods

Results

Discussion

Conclusion