Patient-Level Prediction of Multi-Classification Task at Prostate MRI Based on End-to-End Framework Learning From Diagnostic Logic of Radiologists

Lizhi Shao,Zhenyu Liu,Ye Yan,Jiangang Liu,Xiongjun Ye,Haizhui Xia,Xuehua Zhu,Yuting Zhang,Zhiying Zhang,Huiying Chen,Wei He,Cheng Liu,Min Lu,Yi Huang,Kai Sun,Xuezhi Zhou,Guanyu Yang,Jian Lu,Jie Tian

DOI: https://doi.org/10.1109/tbme.2021.3082176

IF: 4.6

2021-12-01

IEEE Transactions on Biomedical Engineering

Abstract:The grade groups (GGs) of Gleason scores (Gs) is the most critical indicator in the clinical diagnosis and treatment system of prostate cancer. End-to-end method for stratifying the patient-level pathological appearance of prostate cancer (PCa) in magnetic resonance (MRI) are of high demand for clinical decision. Existing methods typically employ a statistical method for integrating slice-level results to a patient-level result, which ignores the asymmetric use of ground truth (GT) and overall optimization. Therefore, more domain knowledge (e.g., diagnostic logic of radiologists) needs to be incorporated into the design of the framework. The patient-level GT is necessary to be logically assigned to each slice of a MRI to achieve joint optimization between slice-level analysis and patient-level decision-making. In this paper, we propose a framework (PCa-GGNet-v2) that learns from radiologists to capture signs in a separate two-dimensional (2-D) space of MRI and further associate them for the overall decision, where all steps are optimized jointly in an end-to-end trainable way. In the training phase, patient-level prediction is transferred from weak supervision to supervision with GT. An association route records the attentional slice for reweighting loss of MRI slices and interpretability. We evaluate our method in an in-house multi-center dataset (N = 570) and PROSTATEx (N = 204), which yields five-classification accuracy over 80% and AUC of 0.804 at patient-level respectively. Our method reveals the state-of-the-art performance for patient-level multi-classification task to personalized medicine.

engineering, biomedical

What problem does this paper attempt to address?

The paper aims to address the grading issue in the clinical diagnosis and treatment of prostate cancer (PCa) patients, particularly in the classification of patient-level pathological phenotypes under magnetic resonance imaging (MRI). Specifically, the paper proposes an end-to-end framework (PCa-GGNet-v2) that learns from radiologists' diagnostic logic to achieve multi-classification tasks for prostate cancer patients. Existing methods typically rely on statistical approaches to integrate slice-level results into patient-level outcomes, which overlooks the asymmetric use of ground truth (GT) and overall optimization. Therefore, it is necessary to incorporate more domain knowledge (such as radiologists' diagnostic logic) into the framework design. The main objectives of the paper include: 1. Designing an end-to-end framework to handle patient-level multi-classification tasks, and this framework can learn from radiologists' diagnostic logic to improve performance and interpretability. 2. Using, for the first time, the grading group (GG-RP) after radical prostatectomy (RP) as the ground truth to predict the pathological grading of prostate cancer patients. 3. Designing a dual-attention network for prostate cancer to better perform pixel-level analysis at the slice level and designing a network based on deep reinforcement learning (DRL) for joint optimization. 4. Proposing a re-weighting slice importance strategy (RWIS) to adjust the importance of different slices during training. 5. Evaluating the proposed framework on a multi-center internal dataset (PCa-GGDATA) and a public dataset (PROSTATEx). Through these contributions, the researchers hope to address the challenges faced by current methods when elevating predictions from the slice level to patient-level decisions, particularly how to quantitatively analyze each MRI slice, eliminate redundant information interference under weak supervision, transform slice-level results into patient-level conclusions, and make the decision process align with the logic of clinical practitioners to enhance interpretability.

Patient-Level Prediction of Multi-Classification Task at Prostate MRI Based on End-to-End Framework Learning From Diagnostic Logic of Radiologists

Patient-level grading prediction of prostate cancer from mp-MRI via GMINet

Radiologist-like artificial intelligence for grade group prediction of radical prostatectomy for reducing upgrading and downgrading from biopsy.

Dual Attention Guided Multiscale Neural Network Trained with Curriculum Learning for Noninvasive Prediction of Gleason Grade Group from MRI.

Deep learning model for the detection of prostate cancer and classification of clinically significant disease using multiparametric MRI in comparison to PI-RADs score

Deep Learning in Prostate Cancer Diagnosis Using Multiparametric Magnetic Resonance Imaging With Whole-Mount Histopathology Referenced Delineations

ProstAttention-Net: A deep attention model for prostate cancer segmentation by aggressiveness in MRI scans

Ensemble Attention Guided Multi-SEANet Trained with Curriculum Learning for Noninvasive Prediction of Gleason Grade Groups from MRI

MRI‐based prostate cancer classification using 3D efficient capsule network

Identifying pathological groups from MRI in prostate cancer using graph representation learning

Prostate Cancer Classification and Interpretation With Multiparametric Magnetic Resonance Imaging and Gleason Grade Score Using DarkNet53 Model

A hybrid classification model with radiomics and CNN for high and low grading of prostate cancer Gleason score on mp-MRI

Attention-guided multi-scale learning network for automatic prostate and tumor segmentation on MRI

Enhanced Cascade Prostate Cancer Classifier in mp-MRI Utilizing Recall Feedback Adaptive Loss and Prior Knowledge-Based Feature Extraction

A radiomics based method for prediction of prostate cancer Gleason score using enlarged region of interest

Multi-parametric MRI Radiomics Analysis with Ensemble Learning for Prostate Lesion Classification.

Mutually communicated model based on multi‐parametric MRI for automated segmentation and classification of prostate cancer

Towards Multi-modality Fusion and Prototype-based Feature Refinement for Clinically Significant Prostate Cancer Classification in Transrectal Ultrasound

Selecting Proper Combination of Mpmri Sequences for Prostate Cancer Classification Using Multi-Input Convolutional Neuronal Network.

Multi-view radiomics and deep learning modeling for prostate cancer detection based on multi-parametric MRI

An attention-based multi-resolution model for prostate whole slide imageclassification and localization