Artificial intelligence‐based detection of lymph node metastases by PET/CT predicts prostate cancer‐specific survival

Abstract

Introduction Lymph node metastases are a key prognostic factor in prostate cancer (PCa), but detecting lymph node lesions from PET/CT images is a subjective process resulting in inter‐reader variability. Artificial intelligence (AI)‐based methods can provide an objective image analysis. We aimed at developing and validating an AI‐based tool for detection of lymph node lesions. Methods A group of 399 patients with biopsy‐proven PCa who had undergone 18F‐choline PET/CT for staging prior to treatment were used to train (n = 319) and test (n = 80) the AI‐based tool. The tool consisted of convolutional neural networks using complete PET/CT scans as inputs. In the test set, the AI‐based lymph node detections were compared to those of two independent readers. The association with PCa‐specific survival was investigated. Results The AI‐based tool detected more lymph node lesions than Reader B (98 vs. 87/117; p = .045) using Reader A as reference. AI‐based tool and Reader A showed similar performance (90 vs. 87/111; p = .63) using Reader B as reference. The number of lymph node lesions detected by the AI‐based tool, PSA, and curative treatment was significantly associated with PCa‐specific survival. Conclusion This study shows the feasibility of using an AI‐based tool for automated and objective interpretation of PET/CT images that can provide assessments of lymph node lesions comparable with that of experienced readers and prognostic information in PCa patients.

Automated analysis of PSMA-PET/CT studies using convolutional neural networks

“…A two-sided sign test was used to analyze the number of cases with increased and decreased variability with support of the AI tool. For lymph nodes the number of detections was analyzed instead of TLU measurements as that number has shown to be associated with survival in our previous work 20 . If an AI detected lymph node has non-zero overlap with a manually detected lymph node it is considered a true positive and otherwise a false negative lymph node.…”

“…It is very difficult to segment lymph nodes directly in CT images. Instead, we trained the AI tool to directly segment lymph nodes in 18F-fluorocholine PET/CT using an organ mask, the CT image and the PET image as input as described in the work by Borrelli et al 20 .…”

“…In a second study AI-based quantification of the primary tumor in the prostate gland showed to be associated with overall survival 19 . A third study showed that AI-based detections of lymph nodes were associated with prostate cancer specific survival 20 . These studies were based on the non-PSMA-targeting PET tracers choline and sodium fluoride for imaging prostate cancer.…”

“…. A third study showed that AI-based detections of lymph nodes were associated with prostate cancer specific survival 20 . These studies were based on the non-PSMA-targeting PET tracers choline and sodium fluoride for imaging prostate cancer.…”

Artificial Intelligence Clinical Application in Prostate Cancer Molecular Imaging

“…The AI-based tool detected more lymph node metastasis than reader B (98 vs. 87; p=0.045 using reader A as reference) and had a similar performance with reader A (90 vs. 87; p=0.63 using reader B as reference). Furthermore, the AI-based number of lymph node metastasis, as well as PSA and curative treatment, were significantly associated with PCspecific survival 53. Based on [ 68 Ga]PSMA11 PET/CT.…”