Multiparametric MRI and Machine Learning Based Radiomic Models for Preoperative Prediction of Multiple Biological Characteristics in Prostate Cancer

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

We developed a novel machine-learning algorithm to augment the clinical diagnosis of prostate cancer utilizing first and second-order texture analysis metrics in a novel application of machine-learning radiomics analysis. We successfully discriminated between significant prostate cancers versus non-tumor regions and provided accurate prediction between Gleason score cohorts with statistical sensitivity of 0.82, 0.81 and 0.91 in three separate pathology classifications. Tumor heterogeneity and prediction of the Gleason score were quantified using two feature selection approaches and two separate classifiers with tuned hyperparameters. There was a total of 71 patients analyzed in this study. Multiparametric MRI, incorporating T2WI and ADC maps, were used to derive radiomics features. Recursive feature elimination (RFE), the least absolute shrinkage and selection operator (LASSO), and two classification approaches, incorporating a support vector machine (SVM) (with randomized search) and random forest (RF) (with grid search), were utilized to differentiate between non-tumor regions and significant cancer while also predicting the Gleason score. In T2WI images, the RFE feature selection approach combined with RF and SVM classifiers outperformed LASSO with SVM and RF classifiers. The best performance was achieved by combining LASSO and SVM into a model that used both T2WI and ADC images. This model had an area under the curve (AUC) of 0.91. Radiomic features computed from ADC and T2WI images were used to predict three groups of Gleason score using two kinds of feature selection methods (RFE and LASSO), RF and SVM classifier models with tuned hyperparameters. Using combined sequences (T2WI and ADC map images) and combined radiomics (1st and GLCM features), LASSO, with a feature selection method with RF, was able to predict G3 with the highest sensitivity at a level AUC of 0.92. To predict G3 for single sequence (T2WI images) using GLCM features, LASSO with SVM achieved the highest sensitivity with an AUC of 0.92.

Radiomics for the identification of extraprostatic extension with prostate MRI: a systematic review and meta-analysis

OBJECTIVES

Extraprostatic extension (EPE) of prostate cancer (PCa) is predicted using clinical nomograms. Incorporating MRI could represent a leap forward, although poor sensitivity and standardization represent unsolved issues. MRI radiomics has been proposed for EPE prediction. The aim of the study was to systematically review the literature and perform a meta-analysis of MRI-based radiomics approaches for EPE prediction.

MATERIALS AND METHODS

Multiple databases were systematically searched for radiomics studies on EPE detection up to June 2022. Methodological quality was appraised according to Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool and radiomics quality score (RQS). The area under the receiver operating characteristic curves (AUC) was pooled to estimate predictive accuracy. A random-effects model estimated overall effect size. Statistical heterogeneity was assessed with I2 value. Publication bias was evaluated with a funnel plot. Subgroup analyses were performed to explore heterogeneity.

RESULTS

Thirteen studies were included, showing limitations in study design and methodological quality (median RQS 10/36), with high statistical heterogeneity. Pooled AUC for EPE identification was 0.80. In subgroup analysis, test-set and cross-validation-based studies had pooled AUC of 0.85 and 0.89 respectively. Pooled AUC was 0.72 for deep learning (DL)-based and 0.82 for handcrafted radiomics studies and 0.79 and 0.83 for studies with multiple and single scanner data, respectively. Finally, models with the best predictive performance obtained using radiomics features showed pooled AUC of 0.82, while those including clinical data of 0.76.

CONCLUSION

MRI radiomics-powered models to identify EPE in PCa showed a promising predictive performance overall. However, methodologically robust, clinically driven research evaluating their diagnostic and therapeutic impact is still needed.

CLINICAL RELEVANCE STATEMENT

Radiomics might improve the management of prostate cancer patients increasing the value of MRI in the assessment of extraprostatic extension. However, it is imperative that forthcoming research prioritizes confirmation studies and a stronger clinical orientation to solidify these advancements.

KEY POINTS

• MRI radiomics deserves attention as a tool to overcome the limitations of MRI in prostate cancer local staging. • Pooled AUC was 0.80 for the 13 included studies, with high heterogeneity (84.7%, p < .001), methodological issues, and poor clinical orientation. • Methodologically robust radiomics research needs to focus on increasing MRI sensitivity and bringing added value to clinical nomograms at patient level.

MRI-based radiomics for prediction of extraprostatic extension of prostate cancer: a systematic review and meta-analysis

PURPOSE

We to systematically evaluate the diagnostic performance of MRI radiomics in detecting extracapsular extension (EPE) of prostate cancer (PCa).

METHODS

A literature search of online databases of PubMed, EMBASE, Cochrane Library, Web of Science, and Google Scholar online scientific publication databases was performed to identify studies published up to July 2023. The summary estimates were pooled with the hierarchical summary receiver-operating characteristic (HSROC) model. This study was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement, the quality of included studies was assessed with the Quality Assessment of Diagnostic Accuracy Studies-2 tool (QUADAS-2) and the radiomics quality score (RQS). Meta-regression and subgroup analyses were performed to explore the impact of varying clinical settings.

RESULTS

A total of ten studies met the inclusion criteria. The pooled sensitivity and specificity were 0.77 (95% CI 0.68-0.84, I2 = 83.5%) and 0.75 (95% CI 0.67-0.82, I2 = 83.5%), respectively, with an area under the HSROC curve of 0.88 (95% CI 0.85-0.91). Study quality was not high while assessing with the RQS. Substantial heterogeneity was observed between studies; however, meta-regression analysis did not reveal any significant contributing factors.

CONCLUSIONS

MRI radiomics demonstrated moderate sensitivity and specificity, offering similar diagnostic performance with previous risk stratifications and models that primarily based on radiologists' subjective experience. However, all studies included were retrospective, thus the performance of radiomics needs to validate in prospective, multicenter studies.

Prediction of prostate cancer aggressiveness using magnetic resonance imaging radiomics: a dual-center study

PURPOSE

The Gleason score (GS) and positive needles are crucial aggressive indicators of prostate cancer (PCa). This study aimed to investigate the usefulness of magnetic resonance imaging (MRI) radiomics models in predicting GS and positive needles of systematic biopsy in PCa.

MATERIAL AND METHODS

A total of 218 patients with pathologically proven PCa were retrospectively recruited from 2 centers. Small-field-of-view high-resolution T2-weighted imaging and post-contrast delayed sequences were selected to extract radiomics features. Then, analysis of variance and recursive feature elimination were applied to remove redundant features. Radiomics models for predicting GS and positive needles were constructed based on MRI and various classifiers, including support vector machine, linear discriminant analysis, logistic regression (LR), and LR using the least absolute shrinkage and selection operator. The models were evaluated with the area under the curve (AUC) of the receiver-operating characteristic.

RESULTS

The 11 features were chosen as the primary feature subset for the GS prediction, whereas the 5 features were chosen for positive needle prediction. LR was chosen as classifier to construct the radiomics models. For GS prediction, the AUC of the radiomics models was 0.811, 0.814, and 0.717 in the training, internal validation, and external validation sets, respectively. For positive needle prediction, the AUC was 0.806, 0.811, and 0.791 in the training, internal validation, and external validation sets, respectively.

CONCLUSIONS

MRI radiomics models are suitable for predicting GS and positive needles of systematic biopsy in PCa. The models can be used to identify aggressive PCa using a noninvasive, repeatable, and accurate diagnostic method.

CT-based radiomics for predicting pathological grade in hepatocellular carcinoma

Objective

To construct and validate radiomics models for hepatocellular carcinoma (HCC) grade predictions based on contrast-enhanced CT (CECT).

Methods

Patients with pathologically confirmed HCC after surgery and underwent CECT at our institution between January 2016 and December 2020 were enrolled and randomly divided into training and validation datasets. With tumor segmentation and feature extraction, radiomic models were constructed using univariate analysis, followed by least absolute shrinkage and selection operator (LASSO) regression. In addition, combined models with clinical factors and radiomics scores (Radscore) were constructed using logistic regression. Finally, all models were evaluated using the receiver operating characteristic (ROC) curve with the area under the curve (AUC), calibration curve, and decision curve analysis (DCA).

Results

In total 242 patients were enrolled in this study, of whom 170 and 72 formed the training and validation datasets, respectively. The arterial phase and portal venous phase (AP+VP) radiomics model were evaluated as the best for predicting HCC pathological grade among all the models built in our study (AUC = 0.981 in the training dataset; AUC = 0.842 in the validation dataset) and was used to build a nomogram. Furthermore, the calibration curve and DCA indicated that the AP+VP radiomics model had a satisfactory prediction efficiency.

Conclusions

Low- and high-grade HCC can be distinguished with good diagnostic performance using a CECT-based radiomics model.

Applying machine-learning models to differentiate benign and malignant thyroid nodules classified as C-TIRADS 4 based on 2D-ultrasound combined with five contrast-enhanced ultrasound key frames

Objectives

To apply machine learning to extract radiomics features from thyroid two-dimensional ultrasound (2D-US) combined with contrast-enhanced ultrasound (CEUS) images to classify and predict benign and malignant thyroid nodules, classified according to the Chinese version of the thyroid imaging reporting and data system (C-TIRADS) as category 4.

Materials and methods

This retrospective study included 313 pathologically diagnosed thyroid nodules (203 malignant and 110 benign). Two 2D-US images and five CEUS key frames ("2nd second after the arrival time" frame, "time to peak" frame, "2nd second after peak" frame, "first-flash" frame, and "second-flash" frame) were selected to manually label the region of interest using the "Labelme" tool. A total of 7 images of each nodule and their annotates were imported into the Darwin Research Platform for radiomics analysis. The datasets were randomly split into training and test cohorts in a 9:1 ratio. Six classifiers, namely, support vector machine, logistic regression, decision tree, random forest (RF), gradient boosting decision tree and extreme gradient boosting, were used to construct and test the models. Performance was evaluated using a receiver operating characteristic curve analysis. The area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy (ACC), and F1-score were calculated. One junior radiologist and one senior radiologist reviewed the 2D-US image and CEUS videos of each nodule and made a diagnosis. We then compared their AUC and ACC with those of our best model.

Results

The AUC of the diagnosis of US, CEUS and US combined CEUS by junior radiologist and senior radiologist were 0.755, 0.750, 0.784, 0.800, 0.873, 0.890, respectively. The RF classifier performed better than the other five, with an AUC of 1 for the training cohort and 0.94 (95% confidence interval 0.88-1) for the test cohort. The sensitivity, specificity, accuracy, PPV, NPV, and F1-score of the RF model in the test cohort were 0.82, 0.93, 0.90, 0.85, 0.92, and 0.84, respectively. The RF model with 2D-US combined with CEUS key frames achieved equivalent performance as the senior radiologist (AUC: 0.94 vs. 0.92, P = 0.798; ACC: 0.90 vs. 0.92) and outperformed the junior radiologist (AUC: 0.94 vs. 0.80, P = 0.039, ACC: 0.90 vs. 0.81) in the test cohort.

Conclusions

Our model, based on 2D-US and CEUS key frames radiomics features, had good diagnostic efficacy for thyroid nodules, which are classified as C-TIRADS 4. It shows promising potential in assisting less experienced junior radiologists.

A dynamic nomogram for predicting intraoperative brain bulge during decompressive craniectomy in patients with traumatic brain injury: a retrospective study

OBJECTIVE

The aim of this paper is to investigate the risk factors associated with intraoperative brain bulge (IOBB), especially the computed tomography (CT) value of the diseased lateral transverse sinus, and to develop a reliable predictive model to alert neurosurgeons to the possibility of IOBB.

METHODS

A retrospective analysis was performed on 937 patients undergoing traumatic decompressive craniectomy. A total of 644 patients from Fuzong Clinical Medical College of Fujian Medical University were included in the development cohort, and 293 patients from the First Affiliated Hospital of Shantou University Medical College were included in the external validation cohort. Univariate and multifactorial logistic regression analyses identified independent risk factors associated with IOBB. The logistic regression models consisted of independent risk factors, and receiver operating characteristic curves, calibration, and decision curve analyses were used to assess the performance of the models. Various machine learning models were used to compare with the logistic regression model and analyze the importance of the factors, which were eventually jointly developed into a dynamic nomogram for predicting IOBB and published online in the form of a simple calculator.

RESULTS

IOBB occurred in 93/644 (14.4%) patients in the developmental cohort and 47/293 (16.0%) in the validation cohort. Univariate and multifactorial regression analyses showed that age, subdural hematoma, contralateral fracture, brain contusion, and CT value of the diseased lateral transverse sinus were associated with IOBB. A logistic regression model (full model) consisting of the above risk factors had excellent predictive power in both the development cohort [area under the curve (AUC)=0.930] and the validation cohort (AUC=0.913). Among the four machine learning models, the AdaBoost model showed the best predictive value (AUC=0.998). Factors in the AdaBoost model were ranked by importance and combined with the full model to create a dynamic nomogram for clinical application, which was published online as a practical and easy-to-use calculator.

CONCLUSIONS

The CT value of the diseased lateral transverse is an independent risk factor and a reliable predictor of IOBB. The online dynamic nomogram formed by combining logistic regression analysis models and machine learning models can more accurately predict the possibility of IOBBs in patients undergoing traumatic decompressive craniectomy.

Multi-parametric MRI-based radiomics for preoperative prediction of multiple biological characteristics in endometrial cancer

Purpose

To develop and validate multi-parametric MRI (MP-MRI)-based radiomics models for the prediction of biological characteristics in endometrial cancer (EC).

Methods

A total of 292 patients with EC were divided into LVSI (n = 208), DMI (n = 292), MSI (n = 95), and Her-2 (n = 198) subsets. Total 2316 radiomics features were extracted from MP-MRI (T2WI, DWI, and ADC) images, and clinical factors (age, FIGO stage, differentiation degree, pathological type, menopausal state, and irregular vaginal bleeding) were included. Intra-class correlation coefficient (ICC), spearman's rank correlation test, univariate logistic regression, and least absolute shrinkage and selection operator (LASSO) were used to select radiomics features; univariate and multivariate logistic regression were used to identify clinical independent risk factors. Five classifiers were applied (logistic regression, random forest, decision tree, K-nearest neighbor, and Bayes) to construct radiomics models for predicting biological characteristics. The clinical model was built based on the clinical independent risk factors. The combined model incorporating the radiomics score (radscore) and the clinical independent risk factors was constructed. The model was evaluated by ROC curve, calibration curve (H-L test), and decision curve analysis (DCA).

Results

In the training cohort, the RF radiomics model performed best among the five classifiers for the three subsets (MSI, LVSI, and DMI) according to AUC values (AUCMSI: 0.844; AUCLVSI: 0.952; AUCDMI: 0.840) except for Her-2 subset (Decision tree: AUC=0.714), and the combined model had higher AUC than the clinical model in each subset (MSI: AUCcombined =0.907, AUCclinical =0.755; LVSI: AUCcombined =0.959, AUCclinical =0.835; DMI: AUCcombined = 0.883, AUCclinical =0.796; Her-2: AUCcombined =0.812, AUCclinical =0.717; all P<0.05). Nevertheless, in the validation cohort, significant differences between the two models (combined vs. clinical model) were found only in the DMI and LVSI subsets (DMI: AUCcombined =0.803, AUCclinical =0.698; LVSI: AUCcombined =0.926, AUCclinical =0.796; all P<0.05).

Conclusion

The radiomics analysis based on MP-MRI and clinical independent risk factors can potentially predict multiple biological features of EC, including DMI, LVSI, MSI, and Her-2, and provide valuable guidance for clinical decision-making.

Value of radiomics-based two-dimensional ultrasound for diagnosing early diabetic nephropathy

Despite efforts to diagnose diabetic nephropathy (DN) using biochemical data or ultrasound imaging separately, a significant gap exists regarding the development of integrated models combining both modalities for enhanced early DN diagnosis. Therefore, we aimed to assess the ability of machine learning models containing two-dimensional ultrasound imaging and biochemical data to diagnose early DN in patients with type 2 diabetes mellitus (T2DM). This retrospective study included 219 patients, divided into a training or test group at an 8:2 ratio. Features were selected using minimum redundancy maximum relevance and random forest-recursive feature elimination. The predictive performance of the models was evaluated using the area under the receiver operating characteristic curve (AUC) for sensitivity, specificity, Matthews Correlation Coefficient, F1 score, and accuracy. K-nearest neighbor, support vector machine, and logistic regression models could diagnose early DN, with AUC values of 0.94, 0.85, and 0.85 in the training cohort and 0.91, 0.84, and 0.84 in the test cohort, respectively. Early DN diagnosing using two-dimensional ultrasound-based radiomics models can potentially revolutionize T2DM patient care by enabling proactive interventions, ultimately improving patient outcomes. Our integrated approach showcases the power of artificial intelligence in medical imaging, enhancing early disease detection strategies with far-reaching applications across medical disciplines.

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

PURPOSE

The Ki67 index and the Gleason grade group (GGG) are vital prognostic indicators of prostate cancer (PCa). This study investigated the value of biparametric magnetic resonance imaging (bpMRI) radiomics feature-based machine learning (ML) models in predicting the Ki67 index and GGG of PCa.

METHODS

A total of 122 patients with pathologically proven PCa who had undergone preoperative MRI were retrospectively included. Radiomics features were extracted from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps. Then, recursive feature elimination (RFE) was applied to remove redundant features. ML models for predicting Ki67 expression and GGG were constructed based on bpMRI and different algorithms, including logistic regression (LR), support vector machine (SVM), random forest (RF), and K-nearest neighbor (KNN). The performances of different models were evaluated with receiver operating characteristic (ROC) analysis. In addition, a joint analysis of Ki67 expression and GGG was performed by assessing their Spearman correlation and calculating the diagnostic accuracy for both indices.

RESULTS

The ML model based on LR and ADC + T2 (LR_ADC + T2, AUC = 0.8882) performed best in predicting Ki67 expression, and ADC_wavelet-LHH_firstorder_Maximum had the highest feature weighting. The SVM_DWI + T2 (AUC = 0.9248) performed best in predicting GGG, and DWI_wavelet HLL_glcm_SumAverage had the highest feature weighting. The Ki67 and GGG exhibited a weak positive correlation (r = 0.382, p < 0.001), and LR_ADC + DWI had the highest diagnostic accuracy in predicting both (0.6230).

CONCLUSION

The proposed ML models are suitable for predicting both Ki67 expression and GGG in PCa. This algorithm could be used to identify indolent or invasive PCa with a noninvasive, repeatable, and accurate diagnostic method.

T2-Weighted MRI Radiomic Features Predict Prostate Cancer Presence and Eventual Biochemical Recurrence

Prostate cancer (PCa) is the most diagnosed non-cutaneous cancer in men. Despite therapies such as radical prostatectomy, which is considered curative, distant metastases may form, resulting in biochemical recurrence (BCR). This study used radiomic features calculated from multi-parametric magnetic resonance imaging (MP-MRI) to evaluate their ability to predict BCR and PCa presence. Data from a total of 279 patients, of which 46 experienced BCR, undergoing MP-MRI prior to surgery were assessed for this study. After surgery, the prostate was sectioned using patient-specific 3D-printed slicing jigs modeled using the T2-weighted imaging (T2WI). Sectioned tissue was stained, digitized, and annotated by a GU-fellowship trained pathologist for cancer presence. Digitized slides and annotations were co-registered to the T2WI and radiomic features were calculated across the whole prostate and cancerous lesions. A tree regression model was fitted to assess the ability of radiomic features to predict BCR, and a tree classification model was fitted with the same radiomic features to classify regions of cancer. We found that 10 radiomic features predicted eventual BCR with an AUC of 0.97 and classified cancer at an accuracy of 89.9%. This study showcases the application of a radiomic feature-based tool to screen for the presence of prostate cancer and assess patient prognosis, as determined by biochemical recurrence.

Deep learning-based radiomic nomograms for predicting Ki67 expression in prostate cancer

BACKGROUND

To explore the value of a multiparametric magnetic resonance imaging (MRI)-based deep learning model for the preoperative prediction of Ki67 expression in prostate cancer (PCa).

MATERIALS

The data of 229 patients with PCa from two centers were retrospectively analyzed and divided into training, internal validation, and external validation sets. Deep learning features were extracted and selected from each patient's prostate multiparametric MRI (diffusion-weighted imaging, T2-weighted imaging, and contrast-enhanced T1-weighted imaging sequences) data to establish a deep radiomic signature and construct models for the preoperative prediction of Ki67 expression. Independent predictive risk factors were identified and incorporated into a clinical model, and the clinical and deep learning models were combined to obtain a joint model. The predictive performance of multiple deep-learning models was then evaluated.

RESULTS

Seven prediction models were constructed: one clinical model, three deep learning models (the DLRS-Resnet, DLRS-Inception, and DLRS-Densenet models), and three joint models (the Nomogram-Resnet, Nomogram-Inception, and Nomogram-Densenet models). The areas under the curve (AUCs) of the clinical model in the testing, internal validation, and external validation sets were 0.794, 0.711, and 0.75, respectively. The AUCs of the deep models and joint models ranged from 0.939 to 0.993. The DeLong test revealed that the predictive performance of the deep learning models and the joint models was superior to that of the clinical model (p < 0.01). The predictive performance of the DLRS-Resnet model was inferior to that of the Nomogram-Resnet model (p < 0.01), whereas the predictive performance of the remaining deep learning models and joint models did not differ significantly.

CONCLUSION

The multiple easy-to-use deep learning-based models for predicting Ki67 expression in PCa developed in this study can help physicians obtain more detailed prognostic data before a patient undergoes surgery.

MRI-based nomograms and radiomics in presurgical prediction of extraprostatic extension in prostate cancer: a systematic review

PURPOSE

Prediction of extraprostatic extension (EPE) is essential for accurate surgical planning in prostate cancer (PCa). Radiomics based on magnetic resonance imaging (MRI) has shown potential to predict EPE. We aimed to evaluate studies proposing MRI-based nomograms and radiomics for EPE prediction and assess the quality of current radiomics literature.

METHODS

We used PubMed, EMBASE, and SCOPUS databases to find related articles using synonyms for MRI radiomics and nomograms to predict EPE. Two co-authors scored the quality of radiomics literature using the Radiomics Quality Score (RQS). Inter-rater agreement was measured using the intraclass correlation coefficient (ICC) from total RQS scores. We analyzed the characteristic s of the studies and used ANOVAs to associate the area under the curve (AUC) to sample size, clinical and imaging variables, and RQS scores.

RESULTS

We identified 33 studies-22 nomograms and 11 radiomics analyses. The mean AUC for nomogram articles was 0.783, and no significant associations were found between AUC and sample size, clinical variables, or number of imaging variables. For radiomics articles, there were significant associations between number of lesions and AUC (p < 0.013). The average RQS total score was 15.91/36 (44%). Through the radiomics operation, segmentation of region-of-interest, selection of features, and model building resulted in a broader range of results. The qualities the studies lacked most were phantom tests for scanner variabilities, temporal variability, external validation datasets, prospective designs, cost-effectiveness analysis, and open science.

CONCLUSION

Utilizing MRI-based radiomics to predict EPE in PCa patients demonstrates promising outcomes. However, quality improvement and standardization of radiomics workflow are needed.

Beyond diagnosis: is there a role for radiomics in prostate cancer management?

The role of imaging in pretreatment staging and management of prostate cancer (PCa) is constantly evolving. In the last decade, there has been an ever-growing interest in radiomics as an image analysis approach able to extract objective quantitative features that are missed by human eye. However, most of PCa radiomics studies have been focused on cancer detection and characterisation. With this narrative review we aimed to provide a synopsis of the recently proposed potential applications of radiomics for PCa with a management-based approach, focusing on primary treatments with curative intent and active surveillance as well as highlighting on recurrent disease after primary treatment. Current evidence is encouraging, with radiomics and artificial intelligence appearing as feasible tools to aid physicians in planning PCa management. However, the lack of external independent datasets for validation and prospectively designed studies casts a shadow on the reliability and generalisability of radiomics models, delaying their translation into clinical practice.Key points• Artificial intelligence solutions have been proposed to streamline prostate cancer radiotherapy planning.• Radiomics models could improve risk assessment for radical prostatectomy patient selection.• Delta-radiomics appears promising for the management of patients under active surveillance.• Radiomics might outperform current nomograms for prostate cancer recurrence risk assessment.• Reproducibility of results, methodological and ethical issues must still be faced before clinical implementation.

Artificial intelligence in multi-parametric magnetic resonance imaging: A review

Multi-parametric magnetic resonance imaging (mpMRI) is an indispensable tool in the clinical workflow for the diagnosis and treatment planning of various diseases. Machine learning-based artificial intelligence (AI) methods, especially those adopting the deep learning technique, have been extensively employed to perform mpMRI image classification, segmentation, registration, detection, reconstruction, and super-resolution. The current availability of increasing computational power and fast-improving AI algorithms have empowered numerous computer-based systems for applying mpMRI to disease diagnosis, imaging-guided radiotherapy, patient risk and overall survival time prediction, and the development of advanced quantitative imaging technology for magnetic resonance fingerprinting. However, the wide application of these developed systems in the clinic is still limited by a number of factors, including robustness, reliability, and interpretability. This survey aims to provide an overview for new researchers in the field as well as radiologists with the hope that they can understand the general concepts, main application scenarios, and remaining challenges of AI in mpMRI. This article is protected by copyright. All rights reserved.