Artificial intelligence-based measurements of PET/CT imaging biomarkers are associated with disease-specific survival of high-risk prostate cancer patients

Abstract

Objective: Artificial intelligence (AI) offers new opportunities for objective quantitative measurements of imaging biomarkers from positron-emission tomography/computed tomography (PET/CT). Clinical image reporting relies predominantly on observer-dependent visual assessment and easily accessible measures like SUV max , representing lesion uptake in a relatively small amount of tissue. Our hypothesis is that measurements of total volume and lesion uptake of the entire tumour would better reflect the diseases activity with prognostic significance, compared with conventional measurements. Methods: An AI-based algorithm was trained to automatically measure the prostate and its tumour content in PET/CT of 145 patients. The algorithm was then tested retrospectively on 285 high-risk patients, who were examined using 18 F-choline PET/CT for primary staging between April 2008 and July 2015. Prostate tumour volume, tumour fraction of the prostate gland, lesion uptake of the entire tumour, and SUV max were obtained automatically. Associations between these measurements, age, PSA, Gleason score and prostate cancer-specific survival were studied, using a Cox proportional-hazards regression model. Results: Twenty-three patients died of prostate cancer during follow-up (median survival 3.8 years). Total tumour volume of the prostate (p ¼ 0.008), tumour fraction of the gland (p ¼ 0.005), total lesion uptake of the prostate (p ¼ 0.02), and age (p ¼ 0.01) were significantly associated with disease-specific survival, whereas SUV max (p ¼ 0.2), PSA (p ¼ 0.2), and Gleason score (p ¼ 0.8) were not. Conclusion: AI-based assessments of total tumour volume and lesion uptake were significantly associated with disease-specific survival in this patient cohort, whereas SUV max and Gleason scores were not. The AI-based approach appears well-suited for clinically relevant patient stratification and monitoring of individual therapy.

Multi-task Bayesian Model Combining FDG-PET/CT Imaging and Clinical Data for Interpretable High-Grade Prostate Cancer Prognosis

“…Alternatively, extraction of handcrafted radiomics (HCR), i.e., predefined quantitative attributes of an image [21], from a manually delineated intraprostatic region of CT and PET images is another standard method [22][23][24]. To obviate the need for manual prostate contouring and facilitate clinical translation, methods based on extraction of HCRs from an automatic segmentation map generated by a convolutional neural network (CNN) [25] have been proposed and validated [26][27][28]. These approaches have inspired the use of deep learning-based radiomics (DLR), i.e., latent vectors in deep layers of a CNN [29], for PCa prognosis [30].…”