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

Kun Luo,Bowen Zheng,Shidong Lv,Jie Tao,Qiang Wei

2024-08-19

Abstract:Prostate cancer is the second most common cancer in males worldwide, and mpMRI is commonly used for diagnosis. However, interpreting mpMRI is challenging and requires expertise from radiologists. This highlights the urgent need for automated grading in mpMRI. Existing studies lack integration of clinical prior information and suffer from uneven training sample distribution due to prevalence. Therefore, we propose a solution that incorporates prior knowledge, addresses the issue of uneven medical sample distribution, and maintains high interpretability in mpMRI. Firstly, we introduce Prior Knowledge-Based Feature Extraction, which mathematically models the PI-RADS criteria for prostate cancer as diagnostic information into model training. Secondly, we propose Adaptive Recall Feedback Loss to address the extremely imbalanced data problem. This method adjusts the training dynamically based on accuracy and recall in the validation set, resulting in high accuracy and recall simultaneously in the testing set.Thirdly, we design an Enhanced Cascade Prostate Cancer Classifier that classifies prostate cancer into different levels in an interpretable way, which refines the classification results and helps with clinical intervention. Our method is validated through experiments on the PI-CAI dataset and outperforms other methods with a more balanced result in both accuracy and recall rate.

Computer Vision and Pattern Recognition,Artificial Intelligence

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to solve several key problems in the automatic grading of prostate cancer multi - parameter magnetic resonance imaging (mpMRI): 1. **Integration of clinical prior information**: - Existing studies lack the integration of clinical prior information (such as the PI - RADS standard) into model training, resulting in limited diagnostic ability of the model for prostate cancer. 2. **Imbalanced sample distribution**: - There are significant differences in the number of samples at different stages of prostate cancer. This will cause the model to be biased towards the majority - class samples, thus affecting the recognition accuracy of the minority - class samples. Especially for specific samples that require a high recall rate, this bias may lead to a serious misdiagnosis risk. 3. **Insufficient model interpretability**: - Although deep - learning models perform well in classification tasks, their black - box nature makes it difficult for doctors to understand the decision - making process of the model, limiting their application in clinical practice. 4. **Handling of complex classification tasks**: - Multi - class classification tasks (such as ISUP grading) are more complex than binary classification tasks, and the performance of existing methods in handling such tasks is not ideal. To solve the above problems, the author proposes an enhanced - cascade prostate cancer classifier, which improves the automatic grading of mpMRI through three main innovations: - **Prior Knowledge - Based Feature Extraction (F - E)**: - Mathematically model the PI - RADS standard as diagnostic information and integrate it into the model training process to improve the generalization ability and interpretability of the model. - **Adaptive Recall Feedback Loss (RFAloss)**: - Dynamically adjust the training process, optimize the loss function according to the accuracy and recall rate feedback on the validation set, to deal with extremely imbalanced data problems while maintaining high accuracy and high recall rate. - **Enhanced Cascade Prostate Cancer Classifier**: - Design a hierarchical classification strategy to gradually refine different levels of prostate cancer and classify in an interpretable way, which is helpful for clinical intervention. These methods work together to enable the model to perform automatic grading more accurately and reliably when processing prostate cancer mpMRI images, and still maintain good performance under a highly imbalanced sample distribution.

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

Prostate cancer inference via weakly-supervised learning using a large collection of negative MRI

Deep Attentive Panoptic Model for Prostate Cancer Detection Using Biparametric MRI Scans

Computer-aided detection of prostate cancer in early stages using multi-parameter MRI: A promising approach for early diagnosis

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

Prostate Cancer Malignancy Detection and localization from mpMRI using auto-Deep Learning: One Step Closer to Clinical Utilization

Implementation of Convolutional Neural Network Architecture on 3D Multiparametric Magnetic Resonance Imaging for Prostate Cancer Diagnosis

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

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

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

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

Differential Diagnosis of Prostate Cancer Grade to Augment Clinical Diagnosis Based on Classifier Models with Tuned Hyperparameters

Evaluation of a Cascaded Deep Learning-based Algorithm for Prostate Lesion Detection at Biparametric MRI

Automatic detection of prostate cancer grades and chronic prostatitis in biparametric MRI

Advancing Prostate Cancer Diagnosis: A Deep Learning Approach for Enhanced Detection in MRI Images

Prostate Lesion Detection and Salient Feature Assessment Using Zone-Based Classifiers

Predicting the Grade of Prostate Cancer Based on a Biparametric MRI Radiomics Signature

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

Cancer Lesion Screening in MRI for Prostate Biopsy

Magnetic resonance imaging radiomics-based prediction of clinically significant prostate cancer in equivocal PI-RADS 3 lesions in the transitional zone

MRI Radiomics-Based Machine Learning Models for Ki67 Expression and Gleason Grade Group Prediction in Prostate Cancer