Free Receiver Operating Characteristic Research Articles

AI-Based Cancer Detection Model for Contrast-Enhanced Mammography.

The recent development of deep neural network models for the analysis of breast images has been a breakthrough in computer-aided diagnostics (CAD). Contrast-enhanced mammography (CEM) is a recent mammography modality providing anatomical and functional imaging of the breast. Despite the clinical benefits it could bring, only a few research studies have been conducted around deep-learning (DL) based CAD for CEM, especially because the access to large databases is still limited. This study presents the development and evaluation of a CEM-CAD for enhancing lesion detection and breast classification. A deep learning enhanced cancer detection model based on a YOLO architecture has been optimized and trained on a large CEM dataset of 1673 patients (7443 images) with biopsy-proven lesions from various hospitals and acquisition systems. The evaluation was conducted using metrics derived from the free receiver operating characteristic (FROC) for the lesion detection and the receiver operating characteristic (ROC) to evaluate the overall breast classification performance. The performances were evaluated for different types of image input and for each patient background parenchymal enhancement (BPE) level. The optimized model achieved an area under the curve (AUROC) of 0.964 for breast classification. Using both low-energy and recombined image as inputs for the DL model shows greater performance than using only the recombined image. For the lesion detection, the model was able to detect 90% of all cancers with a false positive (non-cancer) rate of 0.128 per image. This study demonstrates a high impact of BPE on classification and detection performance. The developed CEM CAD outperforms previously published papers and its performance is comparable to radiologist-reported classification and detection capability.

The Gap in the Thickness: Estimating Effectiveness of Pulmonary Nodule Detection in Thick- and Thin-Section CT Images with 3D Deep Neural Networks

Background and Objectives.There is a noticeable gap in diagnostic evidence strength between the thick and thin scans of Low-Dose CT (LDCT) for pulmonary nodule detection. When the thin scans are needed is unknown, especially when aided with an artificial intelligence nodule detection system.Methods.A case study is conducted with a set of 1,000 pulmonary nodule screening LDCT scans with both thick (5.0mm), and thin (1.0mm) section scans available. Pulmonary nodule detection is performed by human and artificial intelligence models for nodule detection developed using 3D convolutional neural networks (CNNs). The intra-sample consistency is evaluated with thick and thin scans, for both clinical doctor and NN (neural network) models. Free receiver operating characteristic (FROC) is used to measure the accuracy of humans and NNs.Results.Trained NNs outperform humans with small nodules < 6.0mm, which is a good complement to human ability. For nodules > 6.0mm, human and NNs perform similarly while human takes a fractional advantage. By allowing a few more FPs, a significant sensitivity improvement can be achieved with NNs.Conclusions.There is a performance gap between the thick and thin scans for pulmonary nodule detection regarding both false negatives and false positives. NNs can help reduce false negatives when the nodules are small and trade off the false negatives for sensitivity. A combination of human and trained NNs is a promising way to achieve a fast and accurate diagnosis.

Gradual Self-Training via Confidence and Volume Based Domain Adaptation for Multi Dataset Deep Learning-Based Brain Metastases Detection Using Nonlocal Networks on MRI Images.

Research suggests that treatment of multiple brain metastases (BMs) with stereotactic radiosurgery shows improvement when metastases are detected early, providing a case for BM detection capabilities on small lesions. To demonstrate automatic detection of BM on three MRI datasets using a deep learning-based approach. To improve the performance of the network is iteratively co-trained with datasets from different domains. A systematic approach is proposed to prevent catastrophic forgetting during co-training. Retrospective. A total of 156 patients (105 ground truth and 51 pseudo labels) with 1502 BM (BrainMetShare); 121 patients with 722 BM (local); 400 patients with 447 primary gliomas (BrATS). Training/pseudo labels/validation data were distributed 84/51/21 (BrainMetShare). Training/validation data were split: 121/23 (local) and 375/25 (BrATS). A 5 T and 3T/T1 spin-echo postcontrast (T1-gradient echo) (BrainMetShare), 3T/T1 magnetization prepared rapid acquisition gradient echo postcontrast (T1-MPRAGE) (local), 0.5T, 1T, and 1.16 T/T1-weighted-fluid-attenuated inversion recovery (T1-FLAIR) (BrATS). The ground truth was manually segmented by two (BrainMetShare) and four (BrATS) radiologists and manually annotated by one (local) radiologist. Confidence and volume based domain adaptation (CAVEAT) method of co-training the three datasets on a 3D nonlocal convolutional neural network (CNN) architecture was implemented to detect BM. The performance was evaluated using sensitivity and false positive rates per patient (FP/patient) and free receiver operating characteristic (FROC) analysis at seven predefined (1/8, 1/4, 1/2, 1, 2, 4, and 8) FPs per scan. The sensitivity and FP/patient from a held-out set registered 0.811 at 2.952 FP/patient (BrainMetShare), 0.74 at 3.130 (local), and 0.723 at 2.240 (BrATS) using the CAVEAT approach with lesions as small as 1 mm being detected. Improved sensitivities at lower FP can be achieved by co-training datasets via the CAVEAT paradigm to address the problem of data sparsity. 3 TECHNICAL EFFICACY STAGE: 2.

Lymph node detection in MR Lymphography: false positive reduction using multi-view convolutional neural networks.

PurposeTo investigate whether multi-view convolutional neural networks can improve a fully automated lymph node detection system for pelvic MR Lymphography (MRL) images of patients with prostate cancer.MethodsA fully automated computer-aided detection (CAD) system had been previously developed to detect lymph nodes in MRL studies. The CAD system was extended with three types of 2D multi-view convolutional neural networks (CNN) aiming to reduce false positives (FP). A 2D multi-view CNN is an efficient approximation of a 3D CNN, and three types were evaluated: a 1-view, 3-view, and 9-view 2D CNN. The three deep learning CNN architectures were trained and configured on retrospective data of 240 prostate cancer patients that received MRL images as the standard of care between January 2008 and April 2010. The MRL used ferumoxtran-10 as a contrast agent and comprised at least two imaging sequences: a 3D T1-weighted and a 3D T2*-weighted sequence. A total of 5089 lymph nodes were annotated by two expert readers, reading in consensus. A first experiment compared the performance with and without CNNs and a second experiment compared the individual contribution of the 1-view, 3-view, or 9-view architecture to the performance. The performances were visually compared using free-receiver operating characteristic (FROC) analysis and statistically compared using partial area under the FROC curve analysis. Training and analysis were performed using bootstrapped FROC and 5-fold cross-validation.ResultsAdding multi-view CNNs significantly (p < 0.01) reduced false positive detections. The 3-view and 9-view CNN outperformed (p < 0.01) the 1-view CNN, reducing FP from 20.6 to 7.8/image at 80% sensitivity.ConclusionMulti-view convolutional neural networks significantly reduce false positives in a lymph node detection system for MRL images, and three orthogonal views are sufficient. At the achieved level of performance, CAD for MRL may help speed up finding lymph nodes and assessing them for potential metastatic involvement.

Automatic Diagnosing of Suspicious Lesions in Digital Mammograms

Breast cancer is the most common cancer and the leading cause of morbidity and mortality among women’s age between 50 and 74 years across the worldwide. In this paper we’ve proposed a method to detect the suspicious lesions in mammograms, extracting their features and classify them as Normal or Abnormal and Benign or Malignant for diagnosing of breast cancer. This method consists of two major parts: The first one is detection of regions of interest (ROIs). The second one is diagnosing of detected ROIs. This method was tested by Mini Mammography Image Analysis Society (Mini-MIAS) database. To check method’s performance, we’ve used FROC (Free-Receiver Operating Characteristics) curve in the detection part and ROC (Receiver Operating Characteristics) curve in the diagnosis part. Obtained results show that the performance of detection part has sensitivity of 94.27% at 0.67 false positive per image. The performance of diagnosis part has 94.29% accuracy, with 94.11% sensitivity, 94.44% specificity in the classification as normal or abnormal mammogram, and has achieved 94.4%accuracy, with 96.15% sensitivity and 94.54% specificity in the classification as Benign or Malignant mammogram.

Pulmonary nodule detection in CT images based on shape constraint CV model.

Accurate detection of pulmonary nodules remains a technical challenge in computer-aided diagnosis systems because some nodules may adhere to the blood vessels or the lung wall, which have low contrast compared to the surrounding tissues. In this paper, the analysis of typical shape features of candidate nodules based on a shape constraint Chan-Vese (CV) model combined with calculation of the number of blood branches adhered to nodule candidates is proposed to reduce false positive (FP) nodules from candidate nodules. The proposed scheme consists of three major stages: (1) Segmentation of lung parenchyma from computed tomography images. (2) Extraction of candidate nodules. (3) Reduction of FP nodules. A gray level enhancement combined with a spherical shape enhancement filter is introduced to extract the candidate nodules and their sphere-like contour regions. FPs are removed by analysis of the typical shape features of nodule candidates based on the CV model using spherical constraint and by investigating the number of blood branches adhered to the candidate nodules. The constrained shapes of CV model are automatically achieved from the extracted candidate nodules. The detection performance was evaluated on 127 nodules of 103 cases including three types of challenging nodules, which are juxta-pleural nodules, juxta-vascular nodules, and ground glass opacity nodules. The free-receiver operating characteristic (FROC) curve shows that the proposed method is able to detect 88% of all the nodules in the data set with 4 FPs per case. Evaluation shows that the authors' method is feasible and effective for detection of three types of nodules in this study.

Automatic diagnosis of mammographic abnormalities based on hybrid features with learning classifier

Breast cancer screening is currently performed by mammography, which is limited by overlying anatomy and dense breast tissue. Computer-aided detection (CAD) systems can serve as a double reader to improve radiologist performance. In this paper, we have applied a novel approach to segmentation of suspicious region by mammogram and classification based on hybrid features with learning classifier. We formulated differentiation of lesion from normal tissue as a supervised learning problem, and applied this learning method to develop the classification algorithm. The algorithm has been verified with 164 mammograms in the mini Mammographic Image Analysis Society database. The experimental results show that the detection method has a sensitivity of 94.5% at 0.26 false positives per image. The efficiency of algorithm is measured using free receiver operating characteristics curve and the results are highlighted. We conclude that CAD technology with learning classifier has the potential to help radiologists with the task of discriminating between lesion and normal tissues.

Combination of computer-aided detection algorithms for automatic lung nodule identification

The aim of this work is to evaluate the potential of combining different computer-aided detection (CADe) methods to increase the actual support for radiologists of automated systems in the identification of pulmonary nodules in CT scans. The outputs of three different CADe systems developed by researchers of the Italian MAGIC-5 collaboration were combined. The systems are: the CAMCADe (based on a Channeler-Ant-Model which segments vessel tree and nodule candidates and a neural classifier), the RGVPCADe (a Region-Growing- Volume-Plateau algorithm detects nodule candidates and a neural network reduces false positives); the VBNACADe (two dedicated procedures, based respectively on a 3D dot-enhancement algorithm and on intersections of pleura surface normals, identifies internal and juxtapleural nodules, and a Voxel-Based-Neural-Approach reduces false positives. A dedicated OsiriX plugin implemented with the Cocoa environments of MacOSX allows annotating nodules and visualizing singles and combined CADe findings. The combined CADe has been tested on thin slice (lower than 2 mm) CTs of the LIDC public research database and the results have been compared with those obtained by the single systems. The FROC (Free Receiver Operating Characteristic) curves show better results than the best of the single approaches. Has been demonstrated that the combination of different approaches offers better results than each single CADe system. A clinical validation of the combined CADe as second reader is being addressed by means of the dedicated OsiriX plugin.

Finding regions of interest for cancerous masses enhanced by elimination of linear structures and considerations on detection correctness measures in mammography

Cancerous mass detection methods for mammographic images still miss some malignant cases on the one hand, and produce too many false-positive (FP) detections with respect to the number of true-positive (TP) detections on the other. An attempt has been described to improve the TP ratio per image and to decrease the number of FP errors in the hierarchical template matching detector of regions of interest (ROIs) for cancerous masses by eliminating the images of linear structures (LSs) from the mammograms. The LSs were detected with an accumulation-based line detector. The measure of correctness of the ROIs detection was discussed and the quality of the detector, represented by free receiver operating characteristics curves, was compared with the human-eye observations. The result is that the widely used measure of detection correctness seems to underestimate the detection quality made by a human. Tests were performed on the mammograms from the MIAS database.