Multiparametric prostate MRI: technical conduct, standardized report and clinical use

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

Introduction

This study aims to develop an imaging model based on multi-parametric MR images for distinguishing between prostate cancer (PCa) and prostate hyperplasia.

Methods

A total of 236 subjects were enrolled and divided into training and test sets for model construction. Firstly, a multi-view radiomics modeling strategy was designed in which different combinations of radiomics feature categories (original, LoG, and wavelet) were compared to obtain the optimal input feature sets. Minimum-redundancy maximum-relevance (mRMR) selection and least absolute shrinkage selection operator (LASSO) were used for feature reduction, and the next logistic regression method was used for model construction. Then, a Swin Transformer architecture was designed and trained using transfer learning techniques to construct the deep learning models (DL). Finally, the constructed multi-view radiomics and DL models were combined and compared for model selection and nomogram construction. The prediction accuracy, consistency, and clinical benefit were comprehensively evaluated in the model comparison.

Results

The optimal input feature set was found when LoG and wavelet features were combined, while 22 and 17 radiomic features in this set were selected to construct the ADC and T2 multi-view radiomic models, respectively. ADC and T2 DL models were built by transferring learning from a large number of natural images to a relatively small sample of prostate images. All individual and combined models showed good predictive accuracy, consistency, and clinical benefit. Compared with using only an ADC-based model, adding a T2-based model to the combined model would reduce the model's predictive performance. The ADCCombinedScore model showed the best predictive performance among all and was transformed into a nomogram for better use in clinics.

Discussion

The constructed models in our study can be used as a predictor in differentiating PCa and BPH, thus helping clinicians make better clinical treatment decisions and reducing unnecessary prostate biopsies.

Single-Setting 3D MRI/US-Guided Frozen Sectioning and Cryoablation of the Index Lesion: Mid-Term Oncologic and Functional Outcomes from a Pilot Study

Our study explored frozen section reliability in prostate cancer (PCa) diagnoses and described surgical steps of a 3D magnetic resonance imaging (MRI)-ultrasound (US)-guided prostate biopsy (PB) and focal cryoablation of the index lesion (IL) in a single-setting procedure. Patients with a suspicious prostatic specific antigen (PSA) value, with a PIRADS 4 or 5 single lesion, were enrolled for trans perineal 3D MRI-US-guided PB and TRUS-guided focal cryoablation. Three cores were taken from the IL, three cores from the surrounding area, while systematic sampling was performed for the rest of the gland. After confirmation of PCa in frozen sections, focal cryoablation was performed. The 1st-year follow-up schedule included a PSA test at a 3-month interval, MRI 3 months and 1 year postoperatively and PB of the treated area at 1 year. Following the follow-up schedule, an involved PSA test at a 3-month interval and yearly MRI were performed. The PCa diagnosis was histologically confirmed in all three patients with frozen sections. At final histology, a single Gleason score upgrade from 6 (3 + 3) to 7 (3 + 4) was observed. All patients were discharged on postoperative day 1. At the 3-month evaluation, mean PSA values decreased from 12.54 (baseline) to 1.73 ng/mL and MRI images showed complete ablation of the IL in all patients. Urinary continence and potency were preserved in all patients. At the 1-year follow-up, one patient had suspicious ipsilateral recurrence on MRI and underwent a new analogous procedure. Post follow-up was uneventful and PSA remained stable in all patients. Three-dimensional MRI-US-guided frozen sectioning and focal cryoablation of the IL is a step forward towards a "patient-tailored" minimally invasive approach to the diagnosis and cure of PCa.

Magnetic Resonance Features of Acquired Immune Deficiency Syndrome Involving Central Nervous System Diseases by Intelligent Fuzzy C-Means Clustering (FCM) Algorithm

This study was aimed to explore the application of fuzzy C-means (FCM) algorithm in MR images of acquired immune deficiency syndrome (AIDS) patients. Sixty AIDS patients with central nervous disease were selected as the research object. A method of brain MR image segmentation based on FCM clustering optimization was proposed, and FCM was optimized based on the neighborhood pixel correlation of gray difference. The correlation was introduced into the objective function to obtain more accurate pixel membership and segmentation features of the image. The segmented image can retain the original image information. The proposed algorithm can clearly distinguish gray matter from white matter in images. The average time of image segmentation was 0.142 s, the longest time of level set algorithm was 2.887 s, and the running time of multithreshold algorithm was 1.708 s. FCM algorithm had the shortest running time, and the average time was significantly better than other algorithms (P < 0.05). FCM image segmentation efficiency was above 90%, and patients can clearly display the location of lesions after MRI imaging examination. In summary, FCM algorithm can effectively combine the spatial neighborhood information of the brain image, segment the BRAIN MR image, analyze the characteristics of AIDS patients from different directions, and provide effective treatment for patients.

Artificial Intelligence Algorithm-Based MRI for Differentiation Diagnosis of Prostate Cancer

The rapid increase in prostate cancer (PCa) patients is similar to that of benign prostatic hyperplasia (BPH) patients, but the treatments are quite different. In this research, magnetic resonance imaging (MRI) images under the weighted low-rank matrix restoration algorithm (RLRE) were utilized to differentiate PCa from BPH. The diagnostic effects of different sequences of MRI images were evaluated to provide a more effective examination method for the clinical differential diagnosis of PCa and BPH. 150 patients with suspected PCa were taken as the research objects. Pathological examination revealed that 137 patients had PCa and 13 patients had BPH. The pathological results were the gold standard and were compared with the MRI results of different sequences. Therefore, the accuracy of the MRI results was evaluated. The results showed that with the rise of Gaussian noise, the peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) of all three algorithms gradually decreased, but the PSNR and SSIM of the RLRE algorithm were always higher than those of the RL and BM3D algorithms (P < 0.05). The sensitivity (97.08%), specificity (92.31%), accuracy (96.67%), and consistency (0.678) of the dynamic contrast enhancement (DCE) sequence were higher than those of the plain scan (86.13%, 69.23%, 84.67%, and 0.469, respectively). In conclusion, the RLRE algorithm could promote the resolution of MRI images and improve the display effect. DCE could better differentiate PCa from BPH, had great clinical application value, and was worthy of clinical promotion.

Evaluation and Monitoring of Endometrial Cancer Based on Magnetic Resonance Imaging Features of Deep Learning

This study was aimed to compare and analyze the magnetic resonance imaging (MRI) manifestations and surgical pathological results of endometrial cancer (EC) and to explore the clinical research of MRI in the diagnosis and staging of EC. Methods. 80 patients with EC admitted to the hospital were selected as the research objects. The ResNet network was used to optimize the network. When the depth was added, the accuracy of the model was improved, the network parameters were iteratively updated, and the damage function of the minimized network was obtained. The recognition efficiency of MRI images was analyzed using three network modes: shallow CNN network, Res-Net network, and optimized network. The images of EC patients were analyzed, and a quantitative and timed MRI was achieved using simulated datasets in deep learning neural networks, which provided the basis for the formulation of single-scan MRI parameters. All patients underwent preoperative MRI examination using coronal and sagittal T1WI and T2WI imaging. The results showed that the accuracy and specificity of T2 weighted imaging and enhanced scanning in MRI were 88.75% and 95%, respectively. Sensitivity was 87.5%, negative predictive value was 93.75%, and positive predictive value was 86.25%. By MRI examination, 80 cases of EC in patients with stage I diagnosis were 72 cases, accounting for 90%, with endometrial thickening and uneven enhancement. In conclusion, the MRI manifestations of EC are diversified, and MRI has a high value for the staging of EC. MRI examination is conducive to improving diagnostic accuracy.

Diagnostic Value of MRI Combined with CXCR4 Expression Level in Lymph Node Metastasis Head and Neck Squamous Cell Carcinoma

Objective

To explore the diagnostic value of magnetic resonance imaging (MRI) combined with CXCR4 expression levels in lymph node metastasis of the head and neck squamous cell carcinoma (HNSCC).

Methods

289 patients with HNSCC were divided into lymph node metastasis group (LNM group, n = 171) and non-LNM group (n = 118) according to the pathological examination results. MRI was used to scan the patient's lesions and cervical lymph nodes, and ADC was measured by MRI diffusion weighting imaging. The expression of CXCR4 in tumor tissues was detected by qRT-PCR. Logistic regression was used to analyze the risk factors of HNSCC lymph node metastasis. ROC curve was used to analyze the diagnostic effects of MRI, CXCR4, and MRI combined with CXCR4 on HNSCC lymph node metastasis.

Results

Compared with the non-LNM group, patients in the LNM group had a lower degree of pathological differentiation, and the positive rate of TNM staging and vascular invasion was higher. The signal intensity of T1WI and T2WI were low intensity and high intensity, respectively, and the ADC value was significantly reduced. At the same time, the expression level of CXCR4 in the tumor tissues of the LNM group was also significantly increased. In addition, compared with MRI and CXCR4 used alone, MRI combined with CXCR4 has a higher predictive value.

Conclusion

MRI has a good effect in demonstrating lymph node metastasis. CXCR4 is significantly upregulated in lymph node metastasis tumor tissue. The combination of the two can be used for clinical diagnosis of HNSCC lymph node metastasis.

Magnetic Resonance Imaging Image Feature Analysis Algorithm under Convolutional Neural Network in the Diagnosis and Risk Stratification of Prostate Cancer

This work aimed to explore the accuracy of magnetic resonance imaging (MRI) images based on the convolutional neural network (CNN) algorithm in the diagnosis of prostate cancer patients and tumor risk grading. A total of 89 patients with prostate cancer and benign prostatic hyperplasia diagnosed by MRI examination and pathological examination in hospital were selected as the research objects in this study (they passed the exclusion criteria). The MRI images of these patients were collected in two groups and divided into two groups before and after treatment according to whether the CNN algorithm was used to process them. The number of diagnosed diseases and the number of cases of risk level inferred based on the tumor grading were compared to observe which group was closer to the diagnosis of pathological biopsy. Through comparative analysis, compared with the positive rate of pathological diagnosis (44%), the positive rate after the treatment of the CNN algorithm (42%) was more similar to that before the treatment (34%), and the comparison was statistically marked (P < 0.05). In terms of risk stratification, the grading results after treatment (37 cases) were closer to the results of pathological grading (39 cases) than those before treatment (30 cases), and the comparison was statistically obvious (P < 0.05). In addition, it was obvious that the MRT images would be clearer after treatment through the observation of the MRT images before and after treatment. In conclusion, MRI image segmentation algorithm based on CNN was more accurate in the diagnosis and risk stratification of prostate cancer than routine MRI. According to the evaluation of Dice similarity coefficient (DSC) and Hausdorff I distance (HD), the CNN segmentation method used in this study was more perfect than other segmentation methods.

Magnetic Resonance Imaging Sequence Identification Using a Metadata Learning Approach

Despite the wide application of the magnetic resonance imaging (MRI) technique, there are no widely used standards on naming and describing MRI sequences. The absence of consistent naming conventions presents a major challenge in automating image processing since most MRI software require a priori knowledge of the type of the MRI sequences to be processed. This issue becomes increasingly critical with the current efforts toward open-sharing of MRI data in the neuroscience community. This manuscript reports an MRI sequence detection method using imaging metadata and a supervised machine learning technique. Three datasets from the Brain Center for Ontario Data Exploration (Brain-CODE) data platform, each involving MRI data from multiple research institutes, are used to build and test our model. The preliminary results show that a random forest model can be trained to accurately identify MRI sequence types, and to recognize MRI scans that do not belong to any of the known sequence types. Therefore the proposed approach can be used to automate processing of MRI data that involves a large number of variations in sequence names, and to help standardize sequence naming in ongoing data collections. This study highlights the potential of the machine learning approaches in helping manage health data.

The feasibility of a dose painting procedure to treat prostate cancer based on mpMR images and hierarchical clustering

BACKGROUND

We aimed to assess the feasibility of a dose painting (DP) procedure, known as simultaneous integrated boost intensity modulated radiation Therapy (SIB-IMRT), for treating prostate cancer with dominant intraprostatic lesions (DILs) based on multi-parametric magnetic resonance (mpMR) images and hierarchical clustering with a machine learning technique.

METHODS

The mpMR images of 120 patients were used to create hierarchical clustering and draw a dendrogram. Three clusters were selected for performing agglomerative clustering. Then, the DIL acquired from the mpMR images of 20 patients were categorized into three groups to have them treated with a DP procedure being composed of three planning target volumes (PTVs) determined as PTV1, PTV2, and PTV3 in treatment plans. The DP procedure was carried out on the patients wherein a total dose of 80, 85 and 91 Gy were delivered to the PTV1, PTV2, and PTV3, respectively. Dosimetric and radiobiologic parameters [Tumor Control Probability (TCP) and Normal Tissue Complication Probability (NTCP)] of the DP procedure were compared with those of the conventional IMRT and Three-Dimensional Conformal Radiation Therapy (3DCRT) procedures carried out on another group of 20 patients. A post-treatment follow-up was also made four months after the radiotherapy procedures.

RESULTS

All the dosimetric variables and the NTCPs of the organs at risks (OARs) revealed no significant difference between the DP and IMRT procedures. Regarding the TCP of three investigated PTVs, significant differences were observed between the DP versus IMRT and also DP versus 3DCRT procedures. At post-treatment follow-up, the DIL volumes and apparent diffusion coefficient (ADC) values in the DP group differed significantly (p-value < 0.001) from those of the IMRT. However, the whole prostate ADC and prostate-specific antigen (PSA) indicated no significant difference (p-value > 0.05) between the DP versus IMRT.

CONCLUSIONS

The results of this comprehensive clinical trial illustrated the feasibility of our DP procedure for treating prostate cancer based on mpMR images validated with acquired patients' dosimetric and radiobiologic assessment and their follow-ups. This study confirms significant potential of the proposed DP procedure as a promising treatment planning to achieve effective dose escalation and treatment for prostate cancer.

TRIAL REGISTRATION

IRCT20181006041257N1; Iranian Registry of Clinical Trials, Registered: 23 October 2019, https://en.irct.ir/trial/34305 .

Clinical Significance of Color Ultrasound, MRI, miR-21, and CA199 in the Diagnosis of Pancreatic Cancer

Background

To investigate the clinical significance of color ultrasound, magnetic resonance imaging (MRI), miR-21, and CA199 in the diagnosis of pancreatic cancer (PC).

Methods

A total of 160 patients with PC admitted to our hospital from April 2018 to February 2021 were included in the PC group, and another 100 patients with benign pancreatic disease during the same period were included in the pancreatic benign disease group. Color ultrasound and MRI were used for imaging examination of the two groups of PC patients, and the sensitivity, accuracy, and specificity of the two methods for preoperative diagnosis of PC were calculated, respectively. A total of 100 healthy people who underwent physical examination during the same period were included in the control group. Serum CA199 levels of the three groups were detected by ELISA assay. The level of serum miR-21 in the three groups was detected by qRT-PCR. A receiver operating curve (ROC) was drawn to analyze and calculate the sensitivity, specificity, and accuracy of the two serum markers and the combination of color ultrasound and MRI in the detection of PC.

Results

Serum CA199 and miR-21 levels in the PC group were significantly higher than those in the benign lesion group and control group. CA199 and miR-21 levels in the benign lesion group were higher than those in the control group. Both color ultrasound and MRI showed a higher detection rate for PC, and the sensitivity and accuracy were significantly higher than those of CA199 and miR-21. The sensitivity, specificity, and accuracy of combined detection were 91.88%, 96.00%, and 93.46%, respectively, which were significantly higher than those of single detection.

Conclusion

The combined detection of color ultrasound, MRI, miR-21, and CA199 have a high application value in the early diagnosis of PC, which can effectively improve the sensitivity and accuracy of clinical diagnosis, reduce the probability of missed diagnosis and misdiagnosis, and provide a reference for the rational clinical treatment plan and prognosis.

Does mpMRI guidance improve HIFU partial gland ablation compared to conventional ultrasound guidance? Early functional outcomes and complications from a single center

Background

Focal therapy (FT) for localized prostate cancer (PCa) treatment is raising interest. New technological mpMRI-US guided FT devices have never been compared with the previous generation of ultrasound-only guided devices.

Materials and Methods

We retrospectively analyzed prospectively recorded data of men undergoing FT for localized low- or intermediate-risk PCa with US- (Ablatherm®-2009 to 2014) or mpMRI-US (Focal One®-from 2014) guided HIFU. Follow-up visits and data were collected using internationally validated questionnaires at 1, 2, 3, 6 and 12 months.

Results

We included n=88 US-guided FT HIFU and n=52 mpMRI-US guided FT HIFU respectively. No major baseline differences were present except higher rates of Gleason 3+4 for the mpMRI-US group. No major differences were present in hospital stay (p=0.1), catheterization time (p=0.5) and complications (p=0.2) although these tended to be lower in the mpMRI-US group (6.8% versus 13.2% US FT group). At 3 months mpMRI-US guided HIFU had significantly lower urine leak (5.1% vs. 15.9%, p=0.04) and a lower drop in IIEF scores (2 vs. 4.2, p=0.07). Of those undergoing 12-months control biopsy in the mpMRI-US-guided HIFU group, 26% had residual cancer in the treated lobe.

Conclusion

HIFU FT guided by MRI-US fusion may allow improved functional outcomes and fewer complications compared to US- guided HIFU FT alone. Further analysis is needed to confirm benefits of mpMRI implementation at a longer follow-up and on a larger cohort of patients.

Fusion US/MRI prostate biopsy using a computer aided diagnostic (CAD) system

BACKGROUND

The aim of this study was to investigate the impact of computer aided diagnostic (CAD) system on the detection rate of prostate cancer (PCa) in a series of fusion prostate biopsy (FPB).

METHODS

Two prospective transperineal FPB series (with or without CAD assistance) were analyzed and PCa detection rates compared with per-patient and per-target analyses. The χ² and Mann-Whitney test were used to compare categorical and continuous variables, respectively. Univariable and multivariable regression analyses were applied to identify predictors of any and clinically significant (cs) PCa detection. Subgroup analyses were performed after stratifying for PI-RADS Score and lesion location.

RESULTS

Out of 183 FPB, 89 were performed with CAD assistance. At per-patient analysis the detection rate of any PCa and of cs PCa were 56.3% and 30.6%, respectively; the aid of CAD was negligible for either any PCa or csPCa detection rates (P=0.45 and P=0.99, respectively). Conversely in a per-target analysis, CAD-assisted biopsy had significantly higher positive predictive value (PPV) for any PCa versus MRI-only group (58% vs. 37.8%, P=0.001). PI-RADS Score was the only independent predictor of any and csPCa, either in per-patient or per-target multivariable regression analysis (all P<0.029). In a subgroup per-patient analysis of anterior/transitional zone lesions, csPCa detection rate was significantly higher in the CAD cohort (54.5%vs.11.1%, respectively; P=0.028), and CAD assistance was the only predictor of csPCa detection (P=0.013).

CONCLUSIONS

CAD assistance for FPB seems to improve detection of csPCa located in anterior/transitional zone. Enhanced identification and improved contouring of lesions may justify higher diagnostic performance.

Comparison of multiple abbreviated multiparametric MRI-derived protocols for the detection of clinically significant prostate cancer

BACKGROUND

The aim of this paper was to compare the accuracy of multiple abbreviated multiparametric magnetic resonance imaging (mpMRI)-derived protocols in detecting clinically significant prostate cancer (csPCa).

METHODS

One hundred and eight men undergoing staging 3.0T mpMRI with a Prostate Imaging - Reporting and Data System version 2 (PI-RADSv2)-compliant protocol before radical prostatectomy (RP) were retrospectively evaluated. Two readers (R1, R2) independently analyzed mpMRI, assigning a PI-RADSv2 category to each observation as appearing on each examination sequence. A study coordinator assessed final PI-RADSv2 category by combining readers' assignments according to four protocols: short MRI (sMRI) (diffusion-weighted imaging + axial T2-weighted imaging), contrast-enhanced short MRI (cesMRI) (sMRI + dynamic contrast-enhanced [DCE] imaging), biparametric MRI (diffusion-weighted imaging + multiplanar T2-weigthed imaging), and mpMRI. Using RP pathology as the reference standard for csPCa, we calculated the per-lesion cancer detection rate (CDR) and false discovery rate (FDR) for each MRI protocol (cut-off PI-RADSv2 category ≥3), and the per-PI-RADSv2 category prevalence of csPCa and false positives.

RESULTS

Pathology after RP found 142 csPCas with median International Society of Urogenital Pathology grade group 2, and stage ≤pT2c in 68.6% of cases. CDR was comparable across the four MRI protocols (74.6% to 75.3% for R1, and 68.3% for R2). FDR was comparable as well (14.4%-14.5% for R1 and 11.1% for R2). sMRI was the minimum protocol equaling mpMRI in terms of CDR, although cesMRI, similarly to mpMRI, was associated with fewer PI-RADSv2 category 3 assignments and higher prevalence of csPCa within PI-RADSv2 category 3 observations (66.7% versus 76.9% for R1, and 100% versus 91.7% for R2, respectively).

CONCLUSIONS

Among multiple abbreviated mpMRI-derived protocols, cesMRI was the one equaling mpMRI in terms of csPCa detection and minimizing PI-RADSv2 category 3 assignments.