Scholarlink

Background <p>Accurate prediction of clinical outcomes is challenging yet important for patient care. The aim of the study was to evaluate a deep learning-based methodology using tissue-wise information, as a proof of concept, for predicting parameters known to be associated with clinical outcomes.</p> Methods <p>We utilized the publicly available autoPET cohort, consisting of 1014 FDG-PET/CT examinations. Tissue-wise projections were extracted, representing specific tissues (bone, lean tissue, adipose tissue, and air) at different angles. A deep regression and classification framework was trained to predict total metabolic tumor volume (TMTV), lesion count, patient age, sex, and diagnosis status (cancer vs. no cancer). Saliency analysis was performed to identify image regions contributing most to each prediction.</p> Results <p>Here we show that the best model predicts TMTV (MAE = 77 ml; <i>R</i><sup>2</sup> = 0.84; (p &lt;0.05)) and lesion count (MAE = 5.18, <i>R</i><sup>2</sup> = 0.90), when using all tissue-wise projections. Age prediction improves when multiple projection angles are included (MAE = 6.57 years; <i>R</i><sup>2</sup> = 0.70 (p &lt;0.05)). The model also predicts sex (AUC = 1.00 (p &lt;0.05)) and diagnosis status with high accuracy (AUC = 0.95 (p &lt;0.05)).</p> Conclusions <p>This proof-of-concept study demonstrates that tissue-wise projections can be used for efficient and automated prediction of parameters related to clinical outcomes, highlighting their potential for future prediction of clinical outcomes.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Interpretable predictions from whole-body FDG-PET/CT using parameters associated with clinical outcome

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月20日

查看全文
  • Sambit Tarai,
  • Elin Lundström,
  • Nouman Ahmad,
  • Robin Strand,
  • Håkan Ahlström,
  • Joel Kullberg

摘要

Background

Accurate prediction of clinical outcomes is challenging yet important for patient care. The aim of the study was to evaluate a deep learning-based methodology using tissue-wise information, as a proof of concept, for predicting parameters known to be associated with clinical outcomes.

Methods

We utilized the publicly available autoPET cohort, consisting of 1014 FDG-PET/CT examinations. Tissue-wise projections were extracted, representing specific tissues (bone, lean tissue, adipose tissue, and air) at different angles. A deep regression and classification framework was trained to predict total metabolic tumor volume (TMTV), lesion count, patient age, sex, and diagnosis status (cancer vs. no cancer). Saliency analysis was performed to identify image regions contributing most to each prediction.

Results

Here we show that the best model predicts TMTV (MAE = 77 ml; R2 = 0.84; (p <0.05)) and lesion count (MAE = 5.18, R2 = 0.90), when using all tissue-wise projections. Age prediction improves when multiple projection angles are included (MAE = 6.57 years; R2 = 0.70 (p <0.05)). The model also predicts sex (AUC = 1.00 (p <0.05)) and diagnosis status with high accuracy (AUC = 0.95 (p <0.05)).

Conclusions

This proof-of-concept study demonstrates that tissue-wise projections can be used for efficient and automated prediction of parameters related to clinical outcomes, highlighting their potential for future prediction of clinical outcomes.