A Predictive Model for Intraoperative Cerebrospinal Fluid Leak During Endonasal Pituitary Adenoma Resection Using a Convolutional Neural Network.

BACKGROUND: Cerebrospinal fluid (CSF) leak during endoscopic endonasal transsphenoidal surgery can lead to postoperative complications. The clinical and anatomic risk factors of intraoperative CSF leak are not well defined. We applied a two-dimensional (2D) convolutional neural network (CNN) machine learning model to identify risk factors from preoperative magnetic resonance imaging. METHODS: All adults who underwent endoscopic endonasal transsphenoidal surgery at our institution from January 2007 to March 2023 who had accessible preoperative stereotactic magnetic resonance imaging were included. A retrospective classic statistical analysis was performed to identify demographic, clinical, and anatomic risk factors of intraoperative CSF leak. Stereotactic T2-weighted brain magnetic resonance imaging scans were used to train and test a 2D CNN model. RESULTS: Of 220 included patients, 81 (36.8%) experienced intraoperative CSF leak. Among all preoperative variables, visual disturbance was the only statistically significant identified risk factor (P = 0.008). The trained 2D CNN model predicted CSF leak with 92% accuracy and area under receiver operating characteristic curve of 0.90 (sensitivity of 86% and specificity of 93%). Class activation mapping of this model revealed that anatomic regions of CSF flow were most important in predicting CSF leak. CONCLUSIONS: Further review of the class activation mapping gradients revealed regions of the diaphragma sellae, clinoid processes, temporal horns, and optic nerves to have anatomic correlation to intraoperative CSF leak risk. Additionally, visual disturbances from anatomic compression of the optic chiasm were the only identified clinical risk factor. Our 2D CNN model can help a treating team to better anticipate and prepare for intraoperative CSF leak.

Uncovering prostate cancer aggressiveness signal in T2-weighted MRI through a three-reference tissues normalization technique.

Quantitative T2-weighted MRI (T2W) interpretation is impeded by the variability of acquisition-related features, such as field strength, coil type, signal amplification, and pulse sequence parameters. The main purpose of this work is to develop an automated method for prostate T2W intensity normalization. The procedure includes the following: (i) a deep learning-based network utilizing MASK R-CNN for automatic segmentation of three reference tissues: gluteus maximus muscle, femur, and bladder; (ii) fitting a spline function between average intensities in these structures and reference values; and (iii) using the function to transform all T2W intensities. The T2W distributions in the prostate cancer regions of interest (ROIs) and normal appearing prostate tissue (NAT) were compared before and after normalization using Student's t-test. The ROIs' T2W associations with the Gleason Score (GS), Decipher genomic score, and a three-tier prostate cancer risk were evaluated with Spearman's correlation coefficient (rS ). T2W differences in indolent and aggressive prostate cancer lesions were also assessed. The MASK R-CNN was trained with manual contours from 32 patients. The normalization procedure was applied to an independent MRI dataset from 83 patients. T2W differences between ROIs and NAT significantly increased after normalization. T2W intensities in 231 biopsy ROIs were significantly negatively correlated with GS (rS = -0.21, p = 0.001), Decipher (rS = -0.193, p = 0.003), and three-tier risk (rS = -0.235, p < 0.001). The average T2W intensities in the aggressive ROIs were significantly lower than in the indolent ROIs after normalization. In conclusion, the automated triple-reference tissue normalization method significantly improved the discrimination between prostate cancer and normal prostate tissue. In addition, the normalized T2W intensities of cancer exhibited a significant association with tumor aggressiveness. By improving the quantitative utilization of the T2W in the assessment of prostate cancer on MRI, the new normalization method represents an important advance over clinical protocols that do not include sequences for the measurement of T2 relaxation times.

Clinical-Genomic Risk Group Classification of Suspicious Lesions on Prostate Multiparametric-MRI.

The utilization of multi-parametric MRI (mpMRI) in clinical decisions regarding prostate cancer patients' management has recently increased. After biopsy, clinicians can assess risk using National Comprehensive Cancer Network (NCCN) risk stratification schema and commercially available genomic classifiers, such as Decipher. We built radiomics-based models to predict lesions/patients at low risk prior to biopsy based on an established three-tier clinical-genomic classification system. Radiomic features were extracted from regions of positive biopsies and Normally Appearing Tissues (NAT) on T2-weighted and Diffusion-weighted Imaging. Using only clinical information available prior to biopsy, five models for predicting low-risk lesions/patients were evaluated, based on: 1: Clinical variables; 2: Lesion-based radiomic features; 3: Lesion and NAT radiomics; 4: Clinical and lesion-based radiomics; and 5: Clinical, lesion and NAT radiomic features. Eighty-three mpMRI exams from 78 men were analyzed. Models 1 and 2 performed similarly (Area under the receiver operating characteristic curve were 0.835 and 0.838, respectively), but radiomics significantly improved the lesion-based performance of the model in a subset analysis of patients with a negative Digital Rectal Exam (DRE). Adding normal tissue radiomics significantly improved the performance in all cases. Similar patterns were observed on patient-level models. To the best of our knowledge, this is the first study to demonstrate that machine learning radiomics-based models can predict patients' risk using combined clinical-genomic classification.

Generative Adversarial Networks Can Create High Quality Artificial Prostate Cancer Magnetic Resonance Images.

The recent integration of open-source data with machine learning models, especially in the medical field, has opened new doors to studying disease progression and/or regression. However, the ability to use medical data for machine learning approaches is limited by the specificity of data for a particular medical condition. In this context, the most recent technologies, like generative adversarial networks (GANs), are being looked upon as a potential way to generate high-quality synthetic data that preserve the clinical variability of a condition. However, despite some success, GAN model usage remains largely minimal when depicting the heterogeneity of a disease such as prostate cancer. Previous studies from our group members have focused on automating the quantitative multi-parametric magnetic resonance imaging (mpMRI) using habitat risk scoring (HRS) maps on the prostate cancer patients in the BLaStM trial. In the current study, we aimed to use the images from the BLaStM trial and other sources to train the GAN models, generate synthetic images, and validate their quality. In this context, we used T2-weighted prostate MRI images as training data for Single Natural Image GANs (SinGANs) to make a generative model. A deep learning semantic segmentation pipeline trained the model to segment the prostate boundary on 2D MRI slices. Synthetic images with a high-level segmentation boundary of the prostate were filtered and used in the quality control assessment by participating scientists with varying degrees of experience (more than ten years, one year, or no experience) to work with MRI images. Results showed that the most experienced participating group correctly identified conventional vs. synthetic images with 67% accuracy, the group with one year of experience correctly identified the images with 58% accuracy, and the group with no prior experience reached 50% accuracy. Nearly half (47%) of the synthetic images were mistakenly evaluated as conventional. Interestingly, in a blinded quality assessment, a board-certified radiologist did not significantly differentiate between conventional and synthetic images in the context of the mean quality of synthetic and conventional images. Furthermore, to validate the usability of the generated synthetic images from prostate cancer MRIs, we subjected these to anomaly detection along with the original images. Importantly, the success rate of anomaly detection for quality control-approved synthetic data in phase one corresponded to that of the conventional images. In sum, this study shows promise that high-quality synthetic images from MRIs can be generated using GANs. Such an AI model may contribute significantly to various clinical applications which involve supervised machine-learning approaches.

Longitudinal Changes and Predictive Value of Multiparametric MRI Features for Prostate Cancer Patients Treated with MRI-Guided Lattice Extreme Ablative Dose (LEAD) Boost Radiotherapy.

We investigated the longitudinal changes in multiparametric MRI (mpMRI) (T2-weighted, Apparent Diffusion Coefficient (ADC), and Dynamic Contrast Enhanced (DCE-)MRI) of prostate cancer patients receiving Lattice Extreme Ablative Dose (LEAD) radiotherapy (RT) and the capability of their imaging features to predict RT outcome based on endpoint biopsies. Ninety-five mpMRI exams from 25 patients, acquired pre-RT and at 3-, 9-, and 24-months post-RT were analyzed. MRI/Ultrasound-fused biopsies were acquired pre- and at two-years post-RT (endpoint). Five regions of interest (ROIs) were analyzed: Gross tumor volume (GTV), normally-appearing tissue (NAT) and peritumoral volume in both peripheral (PZ) and transition (TZ) zones. Diffusion and perfusion radiomics features were extracted from mpMRI and compared before and after RT using two-tailed Student t-tests. Selected features at the four scan points and their differences (Δ radiomics) were used in multivariate logistic regression models to predict the endpoint biopsy positivity. Baseline ADC values were significantly different between GTV, NAT-PZ, and NAT-TZ (p-values < 0.005). Pharmaco-kinetic features changed significantly in the GTV at 3-month post-RT compared to baseline. Several radiomics features at baseline and three-months post-RT were significantly associated with endpoint biopsy positivity and were used to build models with high predictive power of this endpoint (AUC = 0.98 and 0.89, respectively). Our study characterized the RT-induced changes in perfusion and diffusion. Quantitative imaging features from mpMRI show promise as being predictive of endpoint biopsy positivity.

Deep Learning for Per-Fraction Automatic Segmentation of Gross Tumor Volume (GTV) and Organs at Risk (OARs) in Adaptive Radiotherapy of Cervical Cancer.

Background/Hypothesis: MRI-guided online adaptive radiotherapy (MRI-g-OART) improves target coverage and organs-at-risk (OARs) sparing in radiation therapy (RT). For patients with locally advanced cervical cancer (LACC) undergoing RT, changes in bladder and rectal filling contribute to large inter-fraction target volume motion. We hypothesized that deep learning (DL) convolutional neural networks (CNN) can be trained to accurately segment gross tumor volume (GTV) and OARs both in planning and daily fractions' MRI scans. Materials/Methods: We utilized planning and daily treatment fraction setup (RT-Fr) MRIs from LACC patients, treated with stereotactic body RT to a dose of 45-54 Gy in 25 fractions. Nine structures were manually contoured. MASK R-CNN network was trained and tested under three scenarios: (i) Leave-one-out (LOO), using the planning images of N- 1 patients for training; (ii) the same network, tested on the RT-Fr MRIs of the "left-out" patient, (iii) including the planning MRI of the "left-out" patient as an additional training sample, and tested on RT-Fr MRIs. The network performance was evaluated using the Dice Similarity Coefficient (DSC) and Hausdorff distances. The association between the structures' volume and corresponding DSCs was investigated using Pearson's Correlation Coefficient, r. Results: MRIs from fifteen LACC patients were analyzed. In the LOO scenario the DSC for Rectum, Femur, and Bladder was >0.8, followed by the GTV, Uterus, Mesorectum and Parametrium (0.6-0.7). The results for Vagina and Sigmoid were suboptimal. The performance of the network was similar for most organs when tested on RT-Fr MRI. Including the planning MRI in the training did not improve the segmentation of the RT-Fr MRI. There was a significant correlation between the average organ volume and the corresponding DSC (r = 0.759, p = 0.018). Conclusion: We have established a robust workflow for training MASK R-CNN to automatically segment GTV and OARs in MRI-g-OART of LACC. Albeit the small number of patients in this pilot project, the network was trained to successfully identify several structures while challenges remain, especially in relatively small organs. With the increase of the LACC cases, the performance of the network will improve. A robust auto-contouring tool would improve workflow efficiency and patient tolerance of the OART process.