Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AI Age Discrepancy: A Novel Parameter for Frailty Assessment in Kidney Tumor Patients

Rikhil Seshadri, Jayant Siva, Angelica Bartholomew, Clara Goebel, Gabriel Wallerstein-King, Beatriz López Morato, Nicholas Heller, Jason Scovell, Rebecca Campbell, Andrew Wood, Michal Ozery-Flato, Vesna Barros, Maria Gabrani, Michal Rosen-Zvi, Resha Tejpaul, Vidhyalakshmi Ramesh, Nikolaos Papanikolopoulos, Subodh Regmi, Ryan Ward, Robert Abouassaly, Steven C. Campbell, Erick Remer, Christopher Weight

TL;DR

The paper addresses frailty assessment in kidney tumor patients undergoing surgery by introducing AI Age Discrepancy, a CT-derived frailty metric computed from a ResNet-50 age-prediction model and defined as the normalized residual from predicted to chronological age. The authors demonstrate that AI Age Discrepancy independently associates with longer length of stay ($HR=0.914$, $95\% CI: 0.840-0.994$, $p=0.036$) and worse overall survival ($HR=1.242$, $95\% CI: 1.025-1.504$, $p=0.027$) after adjusting for conventional factors, with additional contributions from lymph node involvement, metastasis, and CCI. This CT-based frailty signal has potential to enhance surgical decision-making and risk stratification, though external validation and prospective studies are needed to establish clinical utility and underlying mechanisms.

Abstract

Kidney cancer is a global health concern, and accurate assessment of patient frailty is crucial for optimizing surgical outcomes. This paper introduces AI Age Discrepancy, a novel metric derived from machine learning analysis of preoperative abdominal CT scans, as a potential indicator of frailty and postoperative risk in kidney cancer patients. This retrospective study of 599 patients from the 2023 Kidney Tumor Segmentation (KiTS) challenge dataset found that a higher AI Age Discrepancy is significantly associated with longer hospital stays and lower overall survival rates, independent of established factors. This suggests that AI Age Discrepancy may provide valuable insights into patient frailty and could thus inform clinical decision-making in kidney cancer treatment.

AI Age Discrepancy: A Novel Parameter for Frailty Assessment in Kidney Tumor Patients

TL;DR

The paper addresses frailty assessment in kidney tumor patients undergoing surgery by introducing AI Age Discrepancy, a CT-derived frailty metric computed from a ResNet-50 age-prediction model and defined as the normalized residual from predicted to chronological age. The authors demonstrate that AI Age Discrepancy independently associates with longer length of stay (, , ) and worse overall survival (, , ) after adjusting for conventional factors, with additional contributions from lymph node involvement, metastasis, and CCI. This CT-based frailty signal has potential to enhance surgical decision-making and risk stratification, though external validation and prospective studies are needed to establish clinical utility and underlying mechanisms.

Abstract

Kidney cancer is a global health concern, and accurate assessment of patient frailty is crucial for optimizing surgical outcomes. This paper introduces AI Age Discrepancy, a novel metric derived from machine learning analysis of preoperative abdominal CT scans, as a potential indicator of frailty and postoperative risk in kidney cancer patients. This retrospective study of 599 patients from the 2023 Kidney Tumor Segmentation (KiTS) challenge dataset found that a higher AI Age Discrepancy is significantly associated with longer hospital stays and lower overall survival rates, independent of established factors. This suggests that AI Age Discrepancy may provide valuable insights into patient frailty and could thus inform clinical decision-making in kidney cancer treatment.
Paper Structure (11 sections, 3 figures, 2 tables)

This paper contains 11 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Dataset
  4. Age Prediction Model
  5. Statistical tests
  6. Results
  7. Length of Hospital Stay
  8. Overall Survival
  9. Discussion
  10. Conclusion
  11. Acknowledgments.

Figures (3)

  • Figure 1: Proposed AI Age Discrepancy Process. ResNet-50 convolutional neural networks are trained to predict age from preoperative CT images. The linear regression line of predictions is calculated. Residuals from predicated age to regression line are taken and then normalized to obtain the AI Age Discrepancy.
  • Figure 2: Scatter plot of patient predicted ages (blue), linear regression line of predicted ages (red), and line of perfect predictions (grey). AI Age Discrepancy is the normalized residual from the predicted age to the linear regression line.
  • Figure 3: Forest plot analysis of (a) LOS and (b) OS. This summarizes the results of the Cox proportional hazards regression to identify factors associated with LOS and OS following kidney cancer surgery. The log(HR) and 95% confidence intervals are presented for each variable. (a) Lower log(HR) values indicate factors associated with a longer LOS. (b) Higher log(HR) values indicate factors associated with an increasing risk of mortality.