Intraobserver Variability in Bladder Cancer Treatment Response Assessment With and Without Computerized Decision Support.

We evaluated the intraobserver variability of physicians aided by a computerized decision-support system for treatment response assessment (CDSS-T) to identify patients who show complete response to neoadjuvant chemotherapy for bladder cancer, and the effects of the intraobserver variability on physicians' assessment accuracy. A CDSS-T tool was developed that uses a combination of deep learning neural network and radiomic features from computed tomography (CT) scans to detect bladder cancers that have fully responded to neoadjuvant treatment. Pre- and postchemotherapy CT scans of 157 bladder cancers from 123 patients were collected. In a multireader, multicase observer study, physician-observers estimated the likelihood of pathologic T0 disease by viewing paired pre/posttreatment CT scans placed side by side on an in-house-developed graphical user interface. Five abdominal radiologists, 4 diagnostic radiology residents, 2 oncologists, and 1 urologist participated as observers. They first provided an estimate without CDSS-T and then with CDSS-T. A subset of cases was evaluated twice to study the intraobserver variability and its effects on observer consistency. The mean areas under the curves for assessment of pathologic T0 disease were 0.85 for CDSS-T alone, 0.76 for physicians without CDSS-T and improved to 0.80 for physicians with CDSS-T (P = .001) in the original evaluation, and 0.78 for physicians without CDSS-T and improved to 0.81 for physicians with CDSS-T (P = .010) in the repeated evaluation. The intraobserver variability was significantly reduced with CDSS-T (P < .0001). The CDSS-T can significantly reduce physicians' variability and improve their accuracy for identifying complete response of muscle-invasive bladder cancer to neoadjuvant chemotherapy.

Variabilities in Reference Standard by Radiologists and Performance Assessment in Detection of Pulmonary Embolism in CT Pulmonary Angiography.

Annotating lesion locations by radiologists' manual marking is a key step to provide reference standard for the training and testing of a computer-aided detection system by supervised machine learning. Inter-reader variability is not uncommon in readings even by expert radiologists. This study evaluated the variability of the radiologist-identified pulmonary emboli (PEs) to demonstrate the importance of improving the reliability of the reference standard by a multi-step process for performance evaluation. In an initial reading of 40 CTPA PE cases, two experienced thoracic radiologists independently marked the PE locations. For markings from the two radiologists that did not agree, each radiologist re-read the cases independently to assess the discordant markings. Finally, for markings that still disagreed after the second reading, the two radiologists read together to reach a consensus. The variability of radiologists was evaluated by analyzing the agreement between two radiologists. For the 40 cases, 475 and 514 PEs were identified by radiologists R1 and R2 in the initial independent readings, respectively. For a total of 545 marks by the two radiologists, 81.5% (444/545) of the marks agreed but 101 marks in 36 cases differed. After consensus, 65 (64.4%) and 36 (35.6%) of the 101 marks were determined to be true PEs and false positives (FPs), respectively. Of these, 48 and 17 were false negatives (FNs) and 14 and 22 were FPs by R1 and R2, respectively. Our study demonstrated that there is substantial variability in reference standards provided by radiologists, which impacts the performance assessment of a lesion detection system. Combination of multiple radiologists' readings and consensus is needed to improve the reliability of a reference standard.

Deep Learning Approach for Assessment of Bladder Cancer Treatment Response.

We compared the performance of different Deep learning-convolutional neural network (DL-CNN) models for bladder cancer treatment response assessment based on transfer learning by freezing different DL-CNN layers and varying the DL-CNN structure. Pre- and posttreatment computed tomography scans of 123 patients (cancers, 129; pre- and posttreatment cancer pairs, 158) undergoing chemotherapy were collected. After chemotherapy 33% of patients had T0 stage cancer (complete response). Regions of interest in pre- and posttreatment scans were extracted from the segmented lesions and combined into hybrid pre -post image pairs (h-ROIs). Training (pairs, 94; h-ROIs, 6209), validation (10 pairs) and test sets (54 pairs) were obtained. The DL-CNN consisted of 2 convolution (C1-C2), 2 locally connected (L3-L4), and 1 fully connected layers. The DL-CNN was trained with h-ROIs to classify cancers as fully responding (stage T0) or not fully responding to chemotherapy. Two radiologists provided lesion likelihood of being stage T0 posttreatment. The test area under the ROC curve (AUC) was 0.73 for T0 prediction by the base DL-CNN structure with randomly initialized weights. The base DL-CNN structure with pretrained weights and transfer learning (no frozen layers) achieved test AUC of 0.79. The test AUCs for 3 modified DL-CNN structures (different C1-C2 max pooling filter sizes, strides, and padding, with transfer learning) were 0.72, 0.86, and 0.69. For the base DL-CNN with (C1) frozen, (C1-C2) frozen, and (C1-C2-L3) frozen, the test AUCs were 0.81, 0.78, and 0.71, respectively. The radiologists' AUCs were 0.76 and 0.77. DL-CNN performed better with pretrained than randomly initialized weights.

Computer-aided assessment of breast density: comparison of supervised deep learning and feature-based statistical learning.

Breast density is one of the most significant factors that is associated with cancer risk. In this study, our purpose was to develop a supervised deep learning approach for automated estimation of percentage density (PD) on digital mammograms (DMs). The input 'for processing' DMs was first log-transformed, enhanced by a multi-resolution preprocessing scheme, and subsampled to a pixel size of 800 µm × 800 µm from 100 µm × 100 µm. A deep convolutional neural network (DCNN) was trained to estimate a probability map of breast density (PMD) by using a domain adaptation resampling method. The PD was estimated as the ratio of the dense area to the breast area based on the PMD. The DCNN approach was compared to a feature-based statistical learning approach. Gray level, texture and morphological features were extracted and a least absolute shrinkage and selection operator was used to combine the features into a feature-based PMD. With approval of the Institutional Review Board, we retrospectively collected a training set of 478 DMs and an independent test set of 183 DMs from patient files in our institution. Two experienced mammography quality standards act radiologists interactively segmented PD as the reference standard. Ten-fold cross-validation was used for model selection and evaluation with the training set. With cross-validation, DCNN obtained a Dice's coefficient (DC) of 0.79 ± 0.13 and Pearson's correlation (r) of 0.97, whereas feature-based learning obtained DC = 0.72 ± 0.18 and r = 0.85. For the independent test set, DCNN achieved DC = 0.76 ± 0.09 and r = 0.94, while feature-based learning achieved DC = 0.62 ± 0.21 and r = 0.75. Our DCNN approach was significantly better and more robust than the feature-based learning approach for automated PD estimation on DMs, demonstrating its potential use for automated density reporting as well as for model-based risk prediction.

Bladder Cancer Segmentation in CT for Treatment Response Assessment: Application of Deep-Learning Convolution Neural Network-A Pilot Study.

Assessing the response of bladder cancer to neoadjuvant chemotherapy is crucial for reducing morbidity and increasing quality of life of patients. Changes in tumor volume during treatment is generally used to predict treatment outcome. We are developing a method for bladder cancer segmentation in CT using a pilot data set of 62 cases. 65 000 regions of interests were extracted from pre-treatment CT images to train a deep-learning convolution neural network (DL-CNN) for tumor boundary detection using leave-one-case-out cross-validation. The results were compared to our previous AI-CALS method. For all lesions in the data set, the longest diameter and its perpendicular were measured by two radiologists, and 3D manual segmentation was obtained from one radiologist. The World Health Organization (WHO) criteria and the Response Evaluation Criteria In Solid Tumors (RECIST) were calculated, and the prediction accuracy of complete response to chemotherapy was estimated by the area under the receiver operating characteristic curve (AUC). The AUCs were 0.73 ± 0.06, 0.70 ± 0.07, and 0.70 ± 0.06, respectively, for the volume change calculated using DL-CNN segmentation, the AI-CALS and the manual contours. The differences did not achieve statistical significance. The AUCs using the WHO criteria were 0.63 ± 0.07 and 0.61 ± 0.06, while the AUCs using RECIST were 0.65 ± 007 and 0.63 ± 0.06 for the two radiologists, respectively. Our results indicate that DL-CNN can produce accurate bladder cancer segmentation for calculation of tumor size change in response to treatment. The volume change performed better than the estimations from the WHO criteria and RECIST for the prediction of complete response.

Treatment response assessment of breast masses on dynamic contrast-enhanced magnetic resonance scans using fuzzy c-means clustering and level set segmentation.

The goal of this study was to develop an automated method to segment breast masses on dynamic contrast-enhanced (DCE) magnetic resonance (MR) scans and to evaluate its potential for estimating tumor volume on pre- and postchemotherapy images and tumor change in response to treatment. A radiologist experienced in interpreting breast MR scans defined a cuboid volume of interest (VOI) enclosing the mass in the MR volume at one time point within the sequence of DCE-MR scans. The corresponding VOIs over the entire time sequence were then automatically extracted. A new 3D VOI representing the local pharmacokinetic activities in the VOI was generated from the 4D VOI sequence by summarizing the temporal intensity enhancement curve of each voxel with its standard deviation. The method then used the fuzzy c-means (FCM) clustering algorithm followed by morphological filtering for initial mass segmentation. The initial segmentation was refined by the 3D level set (LS) method. The velocity field of the LS method was formulated in terms of the mean curvature which guaranteed the smoothness of the surface, the Sobel edge information which attracted the zero LS to the desired mass margin, and the FCM membership function which improved segmentation accuracy. The method was evaluated on 50 DCE-MR scans of 25 patients who underwent neoadjuvant chemotherapy. Each patient had pre- and postchemotherapy DCE-MR scans on a 1.5 T magnet. The in-plane pixel size ranged from 0.546 to 0.703 mm and the slice thickness ranged from 2.5 to 4.5 mm. The flip angle was 15 degrees, repetition time ranged from 5.98 to 6.7 ms, and echo time ranged from 1.2 to 1.3 ms. Computer segmentation was applied to the coronal T1-weighted images. For comparison, the same radiologist who marked the VOI also manually segmented the mass on each slice. The performance of the automated method was quantified using an overlap measure, defined as the ratio of the intersection of the computer and the manual segmentation volumes to the manual segmentation volume. Pre- and postchemotherapy masses had overlap measures of 0.81 +/- 0.13 (mean +/- s.d.) and 0.71 +/- 0.22, respectively. The percentage volume reduction (PVR) estimated by computer and the radiologist were 55.5 +/- 43.0% (mean +/- s.d.) and 57.8 +/- 51.3%, respectively. Paired Student's t test indicated that the difference between the mean PVRs estimated by computer and the radiologist did not reach statistical significance (p = 0.641). The automated mass segmentation method may have the potential to assist physicians in monitoring volume change in breast masses in response to treatment.

ROC study of the effect of stereoscopic imaging on assessment of breast lesions.

An observer performance study was conducted to evaluate the usefulness of assessing breast lesion characteristics with stereomammography. Stereoscopic image pairs of 158 breast biopsy tissue specimens were acquired with a GE Senographe 2000D full field digital mammography system using a 1.8x magnification geometry. A phantom-shift method equivalent to a stereo shift angle of +/- 3 degrees relative to a central axis perpendicular to the detector was used. For each specimen, two pairs of stereo images were taken at approximately orthogonal orientations. The specimens contained either a mass, microcalcifications, both, or normal tissue. Based on pathological analysis, 39.9% of the specimens were found to contain malignancy. The digital specimen radiographs were displayed on a high resolution MegaScan CRT monitor driven by a DOME stereo display board using in-house developed software. Five MQSA radiologists participated as observers. Each observer read the 316 specimen stereo image pairs in a randomized order. For each case, the observer first read the monoscopic image and entered his/her confidence ratings on the presence of microcalcifications and/or masses, margin status, BI-RADS assessment, and the likelihood of malignancy. The corresponding stereoscopic images were then displayed on the same monitor and were viewed through stereoscopic LCD glasses. The observer was free to change the ratings in every category after stereoscopic reading. The ratings of the observers were analyzed by ROC methodology. For the 5 MQSA radiologists, the average Az value for estimation of the likelihood of malignancy of the lesions improved from 0.70 for monoscopic reading to 0.72 (p=0.04) after stereoscopic reading, and the average Az value for the presence of microcalcifications improved from 0.95 to 0.96 (p=0.02). The Az value for the presence of masses improved from 0.80 to 0.82 after stereoscopic reading, but the difference fell short of statistical significance (p=0.08). The visual assessment of margin clearance was found to have very low correlation with microscopic analysis with or without stereoscopic reading. This study demonstrates the potential of using stereomammography to improve the detection and characterization of mammographic lesions.