Massive Training Artificial Neural Network Research Articles

Survey of Machine Learning Applications of Convolutional Neural Networks to Medical Image Analysis

Abstract: Recently, a machine learning (ML) area called deep learning emerged in the computer-vision field and became very popular in many fields. It started from an event in late 2012, when a deep-learning approach based on a convolutional neural network (CNN) won an overwhelming victory in the best-known worldwide computer vision competition, ImageNet Classification. Since then, researchers in many fields, including medical image analysis, have started actively participating in the explosively growing field of deep learning. In this paper, deep learning techniques and their applications to medical image analysis are surveyed. This survey overviewed 1) standard ML techniques in the computer-vision field, 2) what has changed in ML before and after the introduction of deep learning, 3) ML models in deep learning, and 4) applications of deep learning to medical image analysis. The comparisons between MLs before and after deep learning revealed that ML with feature input (or feature-based ML) was dominant before the introduction of deep learning, and that the major and essential difference between ML before and after deep learning is learning image data directly without object segmentation or feature extraction; thus, it is the source of the power of deep learning, although the depth of the model is an important attribute. The survey of deep learningalso revealed that there is a long history of deep-learning techniques in the class of ML with image input, except a new term, “deep learning”. “Deep learning” even before the term existed, namely, the class of ML with image input was applied to various problems in medical image analysis including classification between lesions and nonlesions, classification between lesion types, segmentation of lesions or organs, and detection of lesions. ML with image input including deep learning is a verypowerful, versatile technology with higher performance, which can bring the current state-ofthe-art performance level of medical image analysis to the next level, and it is expected that deep learning will be the mainstream technology in medical image analysis in the next few decades. “Deep learning”, or ML with image input, in medical image analysis is an explosively growing, promising field. It is expected that ML with image input will be the mainstream area in the field of medical image analysis in the next few decades. Keywords: Deep learning, Convolutional neural network, Massive-training artificial neural network, Computer-aided diagnosis, Medical image analysis, Classification (key words)

The Performance of Deep Learning Algorithms on Automatic Pulmonary Nodule Detection and Classification Tested on Different Datasets That Are Not Derived from LIDC-IDRI: A Systematic Review.

The aim of this study was to systematically review the performance of deep learning technology in detecting and classifying pulmonary nodules on computed tomography (CT) scans that were not from the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI) database. Furthermore, we explored the difference in performance when the deep learning technology was applied to test datasets different from the training datasets. Only peer-reviewed, original research articles utilizing deep learning technology were included in this study, and only results from testing on datasets other than the LIDC-IDRI were included. We searched a total of six databases: EMBASE, PubMed, Cochrane Library, the Institute of Electrical and Electronics Engineers, Inc. (IEEE), Scopus, and Web of Science. This resulted in 1782 studies after duplicates were removed, and a total of 26 studies were included in this systematic review. Three studies explored the performance of pulmonary nodule detection only, 16 studies explored the performance of pulmonary nodule classification only, and 7 studies had reports of both pulmonary nodule detection and classification. Three different deep learning architectures were mentioned amongst the included studies: convolutional neural network (CNN), massive training artificial neural network (MTANN), and deep stacked denoising autoencoder extreme learning machine (SDAE-ELM). The studies reached a classification accuracy between 68–99.6% and a detection accuracy between 80.6–94%. Performance of deep learning technology in studies using different test and training datasets was comparable to studies using same type of test and training datasets. In conclusion, deep learning was able to achieve high levels of accuracy, sensitivity, and/or specificity in detecting and/or classifying nodules when applied to pulmonary CT scans not from the LIDC-IDRI database.

Separation of bones from soft tissue in chest radiographs: Anatomy-specific orientation-frequency-specific deep neural network convolution.

Lung nodules that are missed by radiologists as well as by computer-aided detection (CAD) systems mostly overlap with ribs and clavicles. Removing the bony structures would result in better visualization of undetectable lesions. Our purpose in this study was to develop a virtual dual-energy imaging system to separate ribs and clavicles from soft tissue in chest radiographs. We developed a mixture of anatomy-specific, orientation-frequency-specific (ASOFS) deep neural network convolution (NNC) experts. Anatomy-specific (AS) NNC was designed to separate the bony structures from soft tissue in different lung segments. While an AS design was proposed previously under our massive-training artificial neural networks (MTANN) framework, in this work, we newly mathematically defined an AS experts model as well as its learning and inference strategies in a probabilistic deep-learning framework. In addition, in combination with our AS experts design, we newly proposed the orientation-frequency-specific (OFS) NNC models to decompose bone and soft-tissue structures into specific orientation-frequency components of different scales using a multi-resolution decomposition technique. We trained multiple NNC models, each of which is an expert for a specific orientation-frequency component in a particular anatomic segment. Perfect reconstruction discrete wavelet transform was used for OFS decomposition/reconstruction, while we introduced a soft-gating layer to merge the predictions of AS NNC experts. To train our model, we used the bone images obtained from a dual-energy system as the target (or teaching) images while the standard chest radiographs were used as the input to our model. The training, validation, and test were performed in a nested two-fold cross-validation manner. We used a database of 118 chest radiographs with pulmonary nodules to evaluate our NNC scheme. In order to evaluate our scheme, we performed quantitative and qualitative evaluation of the predicted bone and soft-tissue images from our model as well as the ones of a state-of-the-art technique where the "gold-standard" dual-energy bone and soft-tissue images were used as the reference images. Both quantitative and qualitative evaluations demonstrated that our ASOFS NNC was superior to the state-of-the-art bone-suppression technique. Particularly, our scheme was better able to maintain the conspicuity of nodules and lung vessels, comparing to the reference technique, while it separated ribs and clavicles from soft tissue. Comparing to a state-of-the-art bone suppression technique, our bone images had substantially higher (t-test; P<0.01) similarity, in terms of structural similarity index (SSIM) and peak signal-to-noise ratio (PSNR), to the "gold-standard" dual-energy bone images. Our deep ASOFS NNC scheme can decompose chest radiographs into their bone and soft-tissue images accurately, offering the improved conspicuity of lung nodules and vessels, and therefore would be useful for radiologists as well as CAD systems in detecting lung nodules in chest radiographs.

Computer-aided Diagnosis of Lung Cancer in Computed Tomography Scans: A Review

Background: Lung cancer is the main cause of cancer death in both men and women. With the development of Computed Tomography (CT) scans technology, the detection rate of early stage lung cancer has significantly raised. The pulmonary CAD system has been a phenomenal and revolutionary step, in the early and premature detection of lung abnormalities. Discussion: In recent years Computer-Aided Detection (CADe) and Computer-Aided Diagnosis (CADx) system on CT has come to be “the second eyes” of radiologists. It can identify lung nodules on chest CT by itself, and can automatically measure nodule size, density and other information to provide a likelihood of malignancy or benignity. Conclusion: In this article we mainly reviewed the basic method and function of CAD, and its clinical application and research progres. Keywords: Artificial Neural Network (ANN), Computer-Aided Detection system (CADe), Computer-Aided Diagnosis system (CADx), Computed Tomography (CT), Massive Training Artificial Neural Networks (MTANN), Support Vector Machine (SVM).

Robust detection and recognition of traffic signs on road panels

Traffic panels provide vital information on roads with the aid of iconic symbols and text strings. Recognising these signs accurately at the right time is crucial for car drivers to ensure safe journey. The automatic visual recognition and classification of the information contained in the panel could be very useful for driver assistance application. In this paper, a method is proposed to identify and to recognise the information contained in the traffic panels, as an application utilising colour space conversion for image acquisition then the colour decomposition and shape model generation using active appearance model (AAM). The traffic panel is detected using character descriptor-based adaptive fuzzy clustering (AFC). Multiframe detection strategy is a simple way of using precision and recall. Finally the prediction accuracy is computed by using massive training artificial neural network (MTANN). The efficiency of the system is evaluated with the help of the MATLAB-based experimental results which are compared with the bag of visual words for text geolocation and recognising symbols and texts (BTG & ST) method and shape and colour (SC)-based traffic sign detection methods in terms of the sensitivity, specificity and recognition rate.

Robust detection and recognition of traffic signs on road panels

Traffic panels provide vital information on roads with the aid of iconic symbols and text strings. Recognising these signs accurately at the right time is crucial for car drivers to ensure safe journey. The automatic visual recognition and classification of the information contained in the panel could be very useful for driver assistance application. In this paper, a method is proposed to identify and to recognise the information contained in the traffic panels, as an application utilising colour space conversion for image acquisition then the colour decomposition and shape model generation using active appearance model (AAM). The traffic panel is detected using character descriptor-based adaptive fuzzy clustering (AFC). Multiframe detection strategy is a simple way of using precision and recall. Finally the prediction accuracy is computed by using massive training artificial neural network (MTANN). The efficiency of the system is evaluated with the help of the MATLAB-based experimental results which are compared with the bag of visual words for text geolocation and recognising symbols and texts (BTG & ST) method and shape and colour (SC)-based traffic sign detection methods in terms of the sensitivity, specificity and recognition rate.

Overview of deep learning in medical imaging.

The use of machine learning (ML) has been increasing rapidly in the medical imaging field, including computer-aided diagnosis (CAD), radiomics, and medical image analysis. Recently, an ML area called deep learning emerged in the computer vision field and became very popular in many fields. It started from an event in late 2012, when a deep-learning approach based on a convolutional neural network (CNN) won an overwhelming victory in the best-known worldwide computer vision competition, ImageNet Classification. Since then, researchers in virtually all fields, including medical imaging, have started actively participating in the explosively growing field of deep learning. In this paper, the area of deep learning in medical imaging is overviewed, including (1) what was changed in machine learning before and after the introduction of deep learning, (2) what is the source of the power of deep learning, (3) two major deep-learning models: a massive-training artificial neural network (MTANN) and a convolutional neural network (CNN), (4) similarities and differences between the two models, and (5) their applications to medical imaging. This review shows that ML with feature input (or feature-based ML) was dominant before the introduction of deep learning, and that the major and essential difference between ML before and after deep learning is the learning of image data directly without object segmentation or feature extraction; thus, it is the source of the power of deep learning, although the depth of the model is an important attribute. The class of ML with image input (or image-based ML) including deep learning has a long history, but recently gained popularity due to the use of the new terminology, deep learning. There are two major models in this class of ML in medical imaging, MTANN and CNN, which have similarities as well as several differences. In our experience, MTANNs were substantially more efficient in their development, had a higher performance, and required a lesser number of training cases than did CNNs. "Deep learning", or ML with image input, in medical imaging is an explosively growing, promising field. It is expected that ML with image input will be the mainstream area in the field of medical imaging in the next few decades.

Massive-Training Artificial Neural Network Deep Learning in Computer-Aided Diagnosis for Chest and Abdomen

Massive-Training Artificial Neural Network Deep Learning in Computer-Aided Diagnosis for Chest and Abdomen

Comparing two classes of end-to-end machine-learning models in lung nodule detection and classification: MTANNs vs. CNNs

End-to-end learning machines enable a direct mapping from the raw input data to the desired outputs, eliminating the need for hand-crafted features. Despite less engineering effort than the hand-crafted counterparts, these learning machines achieve extremely good results for many computer vision and medical image analysis tasks. Two dominant classes of end-to-end learning machines are massive-training artificial neural networks (MTANNs) and convolutional neural networks (CNNs). Although MTANNs have been actively used for a number of medical image analysis tasks over the past two decades, CNNs have recently gained popularity in the field of medical imaging. In this study, we have compared these two successful learning machines both experimentally and theoretically. For that purpose, we considered two well-studied topics in the field of medical image analysis: detection of lung nodules and distinction between benign and malignant lung nodules in computed tomography (CT). For a thorough analysis, we used 2 optimized MTANN architectures and 4 distinct CNN architectures that have different depths. Our experiments demonstrated that the performance of MTANNs was substantially higher than that of CNN when using only limited training data. With a larger training dataset, the performance gap became less evident even though the margin was still significant. Specifically, for nodule detection, MTANNs generated 2.7 false positives per patient at 100% sensitivity, which was significantly (p<0.05) lower than the best performing CNN model with 22.7 false positives per patient at the same level of sensitivity. For nodule classification, MTANNs yielded an area under the receiver-operating-characteristic curve (AUC) of 0.8806 (95% CI: 0.8389–0.9223), which was significantly (p<0.05) greater than the best performing CNN model with an AUC of 0.7755 (95% CI: 0.7120–0.8270). Thus, with limited training data, MTANNs would be a suitable end-to-end machine-learning model for detection and classification of focal lesions that do not require high-level semantic features.

Separation of Bones From Chest Radiographs by Means of Anatomically Specific Multiple Massive-Training ANNs Combined With Total Variation Minimization Smoothing

Most lung nodules that are missed by radiologists as well as computer-aided detection (CADe) schemes overlap with ribs or clavicles in chest radiographs (CXRs). The purpose of this study was to separate bony structures such as ribs and clavicles from soft tissue in CXRs. To achieve this, we developed anatomically specific multiple massive-training artificial neural networks (MTANNs) combined with total variation (TV) minimization smoothing and a histogram-matching-based consistency improvement method. The anatomically specific multiple MTANNs were designed to separate bones from soft tissue in different anatomic segments of the lungs. Each of the MTANNs was trained with the corresponding anatomic segment in the teaching bone images. The output segmental images from the multiple MTANNs were merged to produce an entire bone image. TV minimization smoothing was applied to the bone image for reduction of noise while preserving edges. This bone image was then subtracted from the original CXR to produce a soft-tissue image where bones were separated out. This new method was compared with conventional MTANNs with a database of 110 CXRs with nodules. Our new anatomically specific MTANNs separated rib edges, ribs close to the lung wall, and the clavicles from soft tissue in CXRs to a substantially higher level than did the conventional MTANNs, while the conspicuity of lung nodules and vessels was maintained. Thus, our technique for bone-soft-tissue separation by means of our new MTANNs would be potentially useful for radiologists as well as CADe schemes in detection of lung nodules on CXRs.

Improved Segmentation of Liver in CT with Massive-Training Artificial Neural Network (MTANN) Liver Enhancer

Improved Segmentation of Liver in CT with Massive-Training Artificial Neural Network (MTANN) Liver Enhancer

Computerized Detection of Lung Nodules by Means of “Virtual Dual-Energy” Radiography

Major challenges in current computer-aided detection (CADe) schemes for nodule detection in chest radiographs (CXRs) are to detect nodules that overlap with ribs and/or clavicles and to reduce the frequent false positives (FPs) caused by ribs. Detection of such nodules by a CADe scheme is very important, because radiologists are likely to miss such subtle nodules. Our purpose in this study was to develop a CADe scheme with improved sensitivity and specificity by use of "virtual dual-energy" (VDE) CXRs where ribs and clavicles are suppressed with massive-training artificial neural networks (MTANNs). To reduce rib-induced FPs and detect nodules overlapping with ribs, we incorporated the VDE technology in our CADe scheme. The VDE technology suppressed rib and clavicle opacities in CXRs while maintaining soft-tissue opacity by use of the MTANN technique that had been trained with real dual-energy imaging. Our scheme detected nodule candidates on VDE images by use of a morphologic filtering technique. Sixty morphologic and gray-level-based features were extracted from each candidate from both original and VDE CXRs. A nonlinear support vector classifier was employed for classification of the nodule candidates. A publicly available database containing 140 nodules in 140 CXRs and 93 normal CXRs was used for testing our CADe scheme. All nodules were confirmed by computed tomography examinations, and the average size of the nodules was 17.8 mm. Thirty percent (42/140) of the nodules were rated "extremely subtle" or "very subtle" by a radiologist. The original scheme without VDE technology achieved a sensitivity of 78.6% (110/140) with 5 (1165/233) FPs per image. By use of the VDE technology, more nodules overlapping with ribs or clavicles were detected and the sensitivity was improved substantially to 85.0% (119/140) at the same FP rate in a leave-one-out cross-validation test, whereas the FP rate was reduced to 2.5 (583/233) per image at the same sensitivity level as the original CADe scheme obtained (Difference between the specificities of the original and the VDE-based CADe schemes was statistically significant). In particular, the sensitivity of our VDE-based CADe scheme for subtle nodules (66.7% = 28/42) was statistically significantly higher than that of the original CADe scheme (57.1% = 24/42). Therefore, by use of VDE technology, the sensitivity and specificity of our CADe scheme for detection of nodules, especially subtle nodules, in CXRs were improved substantially.

Massive‐training support vector regression and Gaussian process for false‐positive reduction in computer‐aided detection of polyps in CT colonography

A massive-training artificial neural network (MTANN) has been developed for the reduction of false positives (FPs) in computer-aided detection (CADe) of polyps in CT colonography (CTC). A major limitation of the MTANN is the long training time. To address this issue, the authors investigated the feasibility of two state-of-the-art regression models, namely, support vector regression (SVR) and Gaussian process regression (GPR) models, in the massive-training framework and developed massive-training SVR (MTSVR) and massive-training GPR (MTGPR) for the reduction of FPs in CADe of polyps. The authors applied SVR and GPR as volume-processing techniques in the distinction of polyps from FP detections in a CTC CADe scheme. Unlike artificial neural networks (ANNs), both SVR and GPR are memory-based methods that store a part of or the entire training data for testing. Therefore, their training is generally fast and they are able to improve the efficiency of the massive-training methodology. Rooted in a maximum margin property, SVR offers excellent generalization ability and robustness to outliers. On the other hand, GPR approaches nonlinear regression from a Bayesian perspective, which produces both the optimal estimated function and the covariance associated with the estimation. Therefore, both SVR and GPR, as the state-of-the-art nonlinear regression models, are able to offer a performance comparable or potentially superior to that of ANN, with highly efficient training. Both MTSVR and MTGPR were trained directly with voxel values from CTC images. A 3D scoring method based on a 3D Gaussian weighting function was applied to the outputs of MTSVR and MTGPR for distinction between polyps and nonpolyps. To test the performance of the proposed models, the authors compared them to the original MTANN in the distinction between actual polyps and various types of FPs in terms of training time reduction and FP reduction performance. The authors' CTC database consisted of 240 CTC data sets obtained from 120 patients in the supine and prone positions. The training set consisted of 27 patients, 10 of which had 10 polyps. The authors selected 10 nonpolyps (i.e., FP sources) from the training set. These ten polyps and ten nonpolyps were used for training the proposed models. The testing set consisted of 93 patients, including 19 polyps in 7 patients and 86 negative patients with 474 FPs produced by an original CADe scheme. With the MTSVR, the training time was reduced by a factor of 190, while a FP reduction performance [by-polyp sensitivity of 94.7% (18/19) with 2.5 (230/93) FPs/patient] comparable to that of the original MTANN [the same sensitivity with 2.6 (244/93) FPs/patient] was achieved. The classification performance in terms of the area under the receiver-operating-characteristic curve value of the MTGPR (0.82) was statistically significantly higher than that of the original MTANN (0.77), with a two-sided p-value of 0.03. The MTGPR yielded a 94.7% (18/19) by-polyp sensitivity at a FP rate of 2.5 (235/93) per patient and reduced the training time by a factor of 1.3. Both MTSVR and MTGPR improve the efficiency of the training in the massive-training framework while maintaining a comparable performance.

Massive-Training Artificial Neural Network Coupled With Laplacian-Eigenfunction-Based Dimensionality Reduction for Computer-Aided Detection of Polyps in CT Colonography

A major challenge in the current computer-aided detection (CAD) of polyps in CT colonography (CTC) is to reduce the number of false-positive (FP) detections while maintaining a high sensitivity level. A pattern-recognition technique based on the use of an artificial neural network (ANN) as a filter, which is called a massive-training ANN (MTANN), has been developed recently for this purpose. The MTANN is trained with a massive number of subvolumes extracted from input volumes together with the teaching volumes containing the distribution for the "likelihood of being a polyp;" hence the term "massive training." Because of the large number of subvolumes and the high dimensionality of voxels in each input subvolume, the training of an MTANN is time-consuming. In order to solve this time issue and make an MTANN work more efficiently, we propose here a dimension reduction method for an MTANN by using Laplacian eigenfunctions (LAPs), denoted as LAP-MTANN. Instead of input voxels, the LAP-MTANN uses the dependence structures of input voxels to compute the selected LAPs of the input voxels from each input subvolume and thus reduces the dimensions of the input vector to the MTANN. Our database consisted of 246 CTC datasets obtained from 123 patients, each of whom was scanned in both supine and prone positions. Seventeen patients had 29 polyps, 15 of which were 5-9 mm and 14 were 10-25 mm in size. We divided our database into a training set and a test set. The training set included 10 polyps in 10 patients and 20 negative patients. The test set had 93 patients including 19 polyps in seven patients and 86 negative patients. To investigate the basic properties of a LAP-MTANN, we trained the LAP-MTANN with actual polyps and a single source of FPs, which were rectal tubes. We applied the trained LAP-MTANN to simulated polyps and rectal tubes. The results showed that the performance of LAP-MTANNs with 20 LAPs was advantageous over that of the original MTANN with 171 inputs. To test the feasibility of the LAP-MTANN, we compared the LAP-MTANN with the original MTANN in the distinction between actual polyps and various types of FPs. The original MTANN yielded a 95% (18/19) by-polyp sensitivity at an FP rate of 3.6 (338/93) per patient, whereas the LAP-MTANN achieved a comparable performance, i.e., an FP rate of 3.9 (367/93) per patient at the same sensitivity level. With the use of the dimension reduction architecture, the time required for training was reduced from 38 h to 4 h. The classification performance in terms of the area under the receiver-operating-characteristic curve of the LAP-MTANN (0.84) was slightly higher than that of the original MTANN (0.82) with no statistically significant difference (p-value =0.48).

CT colonography: Advanced computer‐aided detection scheme utilizing MTANNs for detection of “missed” polyps in a multicenter clinical trial

The purpose of this study was to develop an advanced computer-aided detection (CAD) scheme utilizing massive-training artificial neural networks (MTANNs) to allow detection of "difficult" polyps in CT colonography (CTC) and to evaluate its performance on false-negative (FN) CTC cases that radiologists "missed" in a multicenter clinical trial. The authors developed an advanced CAD scheme consisting of an initial polyp-detection scheme for identification of polyp candidates and a mixture of expert MTANNs for substantial reduction in false positives (FPs) while maintaining sensitivity. The initial polyp-detection scheme consisted of (1) colon segmentation based on anatomy-based extraction and colon-based analysis and (2) detection of polyp candidates based on a morphologic analysis on the segmented colon. The mixture of expert MTANNs consisted of (1) supervised enhancement of polyps and suppression of various types of nonpolyps, (2) a scoring scheme for converting output voxels into a score for each polyp candidate, and (3) combining scores from multiple MTANNs by the use of a mixing artificial neural network. For testing the advanced CAD scheme, they created a database containing 24 FN cases with 23 polyps (range of 6-15 mm; average of 8 mm) and a mass (35 mm), which were "missed" by radiologists in CTC in the original trial in which 15 institutions participated. The initial polyp-detection scheme detected 63% (15/24) of the missed polyps with 21.0 (505/24) FPs per patient. The MTANNs removed 76% of the FPs with loss of one true positive; thus, the performance of the advanced CAD scheme was improved to a sensitivity of 58% (14/24) with 8.6 (207/24) FPs per patient, whereas a conventional CAD scheme yielded a sensitivity of 25% at the same FP rate (the difference was statistically significant). With the advanced MTANN CAD scheme, 58% of the polyps missed by radiologists in the original trial were detected and with a reasonable number of FPs. The results suggest that the use of an advanced MTANN CAD scheme may potentially enhance the detection of "difficult" polyps.

Performance of Radiologists in Detection of Small Pulmonary Nodules on Chest Radiographs: Effect of Rib Suppression With a Massive-Training Artificial Neural Network

A massive-training artificial neural network is a nonlinear pattern recognition tool used to suppress rib opacity on chest radiographs while soft-tissue contrast is maintained. We investigated the effect of rib suppression with a massive-training artificial neural network on the performance of radiologists in the detection of pulmonary nodules on chest radiographs. We used 60 chest radiographs; 30 depicted solitary pulmonary nodules, and 30 showed no nodules. A stratified random-sampling scheme was used to select the images from the standard digital image database developed by the Japanese Society of Radiologic Technology. The mean diameter of the 30 pulmonary nodules was 14.7 +/- 4.1 (SD) mm. Receiver operating characteristic analysis was used to evaluate observer performance in the detection of pulmonary nodules first on the chest radiographs without and then on the radiographs with rib suppression. Seven board-certified radiologists and five radiology residents participated in this observer study. For all 12 observers, the mean values of the area under the best-fit receiver operating characteristic curve for images without and with rib suppression were 0.816 +/- 0.077 and 0.843 +/- 0.074; the difference was statistically significant (p = 0.019). The mean areas under the curve for images without and with rib suppression were 0.848 +/- 0.059 and 0.883 +/- 0.050 for the seven board-certified radiologists (p = 0.011) and 0.770 +/- 0.081 and 0.788 +/- 0.074 for the five radiology residents (p = 0.310). In the detection of pulmonary nodules, evaluation of a combination of rib-suppressed and original chest radiographs significantly improved the diagnostic performance of radiologists over the use of chest radiographs alone.

A supervised ‘lesion-enhancement’ filter by use of a massive-training artificial neural network (MTANN) in computer-aided diagnosis (CAD)**First presented at the Seventh International Conference on Machine Learning and Applications, San Diego, CA, USA, 11–13 December 2008.

Computer-aided diagnosis (CAD) has been an active area of study in medical image analysis. A filter for the enhancement of lesions plays an important role for improving the sensitivity and specificity in CAD schemes. The filter enhances objects similar to a model employed in the filter; e.g. a blob-enhancement filter based on the Hessian matrix enhances sphere-like objects. Actual lesions, however, often differ from a simple model; e.g. a lung nodule is generally modeled as a solid sphere, but there are nodules of various shapes and with internal inhomogeneities such as a nodule with spiculations and ground-glass opacity. Thus, conventional filters often fail to enhance actual lesions. Our purpose in this study was to develop a supervised filter for the enhancement of actual lesions (as opposed to a lesion model) by use of a massive-training artificial neural network (MTANN) in a CAD scheme for detection of lung nodules in CT. The MTANN filter was trained with actual nodules in CT images to enhance actual patterns of nodules. By use of the MTANN filter, the sensitivity and specificity of our CAD scheme were improved substantially. With a database of 69 lung cancers, nodule candidate detection by the MTANN filter achieved a 97% sensitivity with 6.7 false positives (FPs) per section, whereas nodule candidate detection by a difference-image technique achieved a 96% sensitivity with 19.3 FPs per section. Classification-MTANNs were applied for further reduction of the FPs. The classification-MTANNs removed 60% of the FPs with a loss of one true positive; thus, it achieved a 96% sensitivity with 2.7 FPs per section. Overall, with our CAD scheme based on the MTANN filter and classification-MTANNs, an 84% sensitivity with 0.5 FPs per section was achieved.

WE-C-332-09: Effect of Massive-Training ANNs On the Performance of a CAD System On “missed” Polyps in CT Colonography

Purpose: A major challenge in computer‐aided detection (CAD) of polyps in CT colonography (CTC) is the detection of “difficult” polyps which radiologists are likely to miss. Our purpose was to develop a CAD system incorporating massive‐training artificial neural networks (MTANNs) for the detection of polyps and evaluate its performance on polyps “missed” by radiologists in a large multicenter clinical trial. Method and Materials: Our database consists of CTC scans obtained from a multicenter clinical trial in which 15 leading medical institutions participated nationwide. For testing our CAD system, we randomly selected 14 polyps/masses in 14 patients that were “missed” by radiologists in the trial. Lesion sizes ranged from 6–35 mm, with an average of 10 mm. We developed a CAD system consisting of 1) colon segmentation based on CT value‐based analysis and colon tracing, 2) detection of polyp candidates based on morphologic analysis on the segmented colon, 3) calculation of 3D pattern features of the polyp candidates, and 4) classification of the polyp candidates as polyps or non‐polyps by use of a classifier. We tested two different classifiers for step 4): quadratic discriminant analysis based on the pattern features and our original 3D MTANNs. Results: With our CAD system based on quadratic discriminant analysis, we achieved 43% (6/14) by‐patient sensitivity with 5.6 (78/14) false positive detections (FPs) per patient for the 14 “missed” polyps in a leave‐one‐lesion‐out test. We replaced quadratic discriminant analysis with 3D MTANNs, and obtained 71.4% (10/14) sensitivity with 4.8 (67/14) FPs per patient. Conclusion: The performance of 3D MTANNs was superior to that of quadratic discriminant analysis. With our CAD system based on the 3D MTANNs, 71.4% of polyps “missed” by radiologists in the multicenter clinical trial were detected correctly, with an FP rate of 4.8 FPs per patient.

Mixture of expert 3D massive-training ANNs for reduction of multiple types of false positives in CAD for detection of polyps in CT colonography

One of the major challenges in computer-aided detection (CAD) of polyps in CT colonography (CTC) is the reduction of false-positive detections (FPs) without a concomitant reduction in sensitivity. A large number of FPs is likely to confound the radiologist's task of image interpretation, lower the radiologist's efficiency, and cause radiologists to lose their confidence in CAD as a useful tool. Major sources of FPs generated by CAD schemes include haustral folds, residual stool, rectal tubes, the ileocecal valve, and extra-colonic structures such as the small bowel and stomach. Our purpose in this study was to develop a method for the removal of various types of FPs in CAD of polyps while maintaining a high sensitivity. To achieve this, we developed a "mixture of expert" three-dimensional (3D) massive-training artificial neural networks (MTANNs) consisting of four 3D MTANNs that were designed to differentiate between polyps and four categories of FPs: (1) rectal tubes, (2) stool with bubbles, (3) colonic walls with haustral folds, and (4) solid stool. Each expert 3D MTANN was trained with examples from a specific non-polyp category along with typical polyps. The four expert 3D MTANNs were combined with a mixing artificial neural network (ANN) such that different types of FPs could be removed. Our database consisted of 146 CTC datasets obtained from 73 patients whose colons were prepared by standard pre-colonoscopy cleansing. Each patient was scanned in both supine and prone positions. Radiologists established the locations of polyps through the use of optical-colonoscopy reports. Fifteen patients had 28 polyps, 15 of which were 5-9 mm and 13 were 10-25 mm in size. The CTC cases were subjected to our previously reported CAD method consisting of centerline-based extraction of the colon, shape-based detection of polyp candidates, and a Bayesian-ANN-based classification of polyps. The original CAD method yielded 96.4% (27/28) by-polyp sensitivity with an average of 3.1 (224/73) FPs per patient. The mixture of expert 3D MTANNs removed 63% (142/224) of the FPs without the loss of any true positive; thus, the FP rate of our CAD scheme was improved to 1.1 (82/73) FPs per patient while the original sensitivity was maintained. By use of the mixture of expert 3D MTANNs, the specificity of a CAD scheme for detection of polyps in CTC was substantially improved while a high sensitivity was maintained.

Massive‐training artificial neural network (MTANN) for reduction of false positives in computer‐aided detection of polyps: Suppression of rectal tubes

One of the limitations of the current computer-aided detection (CAD) of polyps in CT colonography (CTC) is a relatively large number of false-positive (FP) detections. Rectal tubes (RTs) are one of the typical sources of FPs because a portion of a RT, especially a portion of a bulbous tip, often exhibits a cap-like shape that closely mimics the appearance of a small polyp. Radiologists can easily recognize and dismiss RT-induced FPs; thus, they may lose their confidence in CAD as an effective tool if the CAD scheme generates such "obvious" FPs due to RTs consistently. In addition, RT-induced FPs may distract radiologists from less common true positives in the rectum. Therefore, removal RT-induced FPs as well as other types of FPs is desirable while maintaining a high sensitivity in the detection of polyps. We developed a three-dimensional (3D) massive-training artificial neural network (MTANN) for distinction between polyps and RTs in 3D CTC volumetric data. The 3D MTANN is a supervised volume-processing technique which is trained with input CTC volumes and the corresponding "teaching" volumes. The teaching volume for a polyp contains a 3D Gaussian distribution, and that for a RT contains zeros for enhancement of polyps and suppression of RTs, respectively. For distinction between polyps and nonpolyps including RTs, a 3D scoring method based on a 3D Gaussian weighting function is applied to the output of the trained 3D MTANN. Our database consisted of CTC examinations of 73 patients, scanned in both supine and prone positions (146 CTC data sets in total), with optical colonoscopy as a reference standard for the presence of polyps. Fifteen patients had 28 polyps, 15 of which were 5-9 mm and 13 were 10-25 mm in size. These CTC cases were subjected to our previously reported CAD scheme that included centerline-based segmentation of the colon, shape-based detection of polyps, and reduction of FPs by use of a Bayesian neural network based on geometric and texture features. Application of this CAD scheme yielded 96.4% (27/28) by-polyp sensitivity with 3.1 (224/73) FPs per patient, among which 20 FPs were caused by RTs. To eliminate the FPs due to RTs and possibly other normal structures, we trained a 3D MTANN with ten representative polyps and ten RTs, and applied the trained 3D MTANN to the above CAD true- and false-positive detections. In the output volumes of the 3D MTANN, polyps were represented by distributions of bright voxels, whereas RTs and other normal structures partly similar to RTs appeared as darker voxels, indicating the ability of the 3D MTANN to suppress RTs as well as other normal structures effectively. Application of the 3D MTANN to the CAD detections showed that the 3D MTANN eliminated all RT-induced 20 FPs, as well as 53 FPs due to other causes, without removal of any true positives. Overall, the 3D MTANN was able to reduce the FP rate of the CAD scheme from 3.1 to 2.1 FPs per patient (33% reduction), while the original by-polyp sensitivity of 96.4% was maintained.