A statistical 3-D pattern processing method for computer-aided detection of polyps in CT colonography

Identification of major depression patients using machine learning models based on heart rate variability during sleep stages for pre-hospital screening

With the COVID-19 pandemic causing challenges in hospital admissions globally, the role of home health monitoring in aiding the diagnosis of mental health disorders has become increasingly important. This paper proposes an interpretable machine learning solution to optimise initial screening for major depressive disorder (MDD) in both male and female patients. The data is from the Stanford Technical Analysis and Sleep Genome Study (STAGES). We analyzed 5-min short-term electrocardiogram (ECG) signals during nighttime sleep stages of 40 MDD patients and 40 healthy controls, with a 1:1 gender ratio. After preprocessing, we calculated the time-frequency parameters of heart rate variability (HRV) based on the ECG signals and used common machine learning algorithms for classification, along with feature importance analysis for global decision analysis. Ultimately, the Bayesian optimised extremely randomized trees classifier (BO-ERTC) showed the best performance on this dataset (accuracy 86.32%, specificity 86.49%, sensitivity 85.85%, F1-score 0.86). By using feature importance analysis on the cases confirmed by BO-ERTC, we found that gender is one of the most important factors affecting the prediction of the model, which should not be overlooked in our assisted diagnosis. This method can be embedded in portable ECG monitoring systems and is consistent with the literature results.

Lymph node detection in CT scans using modified U-Net with residual learning and 3D deep network

PURPOSE

Lymph node (LN) detection is a crucial step that complements the diagnosis and treatments involved during cancer investigations. However, the low-contrast structures in the CT scan images and the nodes' varied shapes, sizes, and poses, along with their sparsely distributed locations, make the detection step challenging and lead to many false positives. The manual examination of the CT scan slices could be time-consuming, and false positives could divert the clinician's focus. To overcome these issues, our work aims at providing an automated framework for LNs detection in order to obtain more accurate detection results with low false positives.

METHODS

The proposed work consists of two stages: candidate generation and false positive reduction. The first stage generates volumes of interest (VOI) of probable LN candidates using a modified U-Net with ResNet architecture to obtain high sensitivity but with the cost of increased false positives. The second-stage processes the obtained candidate LNs for false positive reduction using 3D convolutional neural network (CNN) classifier. We further present an analysis of various deep learning models while decomposing 3D VOI into different representations.

RESULTS

The method is evaluated on two publicly available datasets containing CT scans of mediastinal and abdominal LNs. Our proposed approach yields sensitivities of 87% at 2.75 false positives per volume (FP/vol.) and 79% at 1.74 FP/vol. with the mediastinal and abdominal datasets, respectively. Our method presented a competitive performance in terms of sensitivity compared to the state-of-the-art methods and encountered very few false positives.

CONCLUSION

We developed an automated framework for LNs detection using a modified U-Net with residual learning and 3D CNNs. The results indicate that our method could achieve high sensitivity with relatively low false positives, which helps avoid ineffective treatments.

A Survey on Machine Learning and Deep Learning-based Computer-Aided Methods for Detection of Polyps in CT Colonography

BACKGROUND

Colon cancer generally begins as a neoplastic growth of tissue, called polyps, originating from the inner lining of the colon wall. Most colon polyps are considered harmless but over the time, they can evolve into colon cancer, which, when diagnosed in later stages, is often fatal. Hence, time is of the essence in the early detection of polyps and the prevention of colon cancer.

METHODS

To aid this endeavor, many computer-aided methods have been developed, which use a wide array of techniques to detect, localize and segment polyps from CT Colonography images. In this paper, a comprehensive state-of-the-art method is proposed and categorize this work broadly using the available classification techniques using Machine Learning and Deep Learning.

CONCLUSION

The performance of each of the proposed approach is analyzed with existing methods and also how they can be used to tackle the timely and accurate detection of colon polyps.

Grading of hepatocellular carcinoma based on diffusion weighted images with multiple b-values using convolutional neural networks

PURPOSE

To effectively grade hepatocellular carcinoma (HCC) based on deep features derived from diffusion weighted images (DWI) with multiple b-values using convolutional neural networks (CNN).

MATERIALS AND METHODS

Ninety-eight subjects with 100 pathologically confirmed HCC lesions from July 2012 to October 2018 were included in this retrospective study, including 47 low-grade and 53 high-grade HCCs. DWI was performed for each subject with a 3.0T MR scanner in a breath-hold routine with three b-values (0,100, and 600 s/mm² ). First, logarithmic transformation was performed on original DWI images to generate log maps (logb0, logb100, and logb600). Then, a resampling method was performed to extract multiple 2D axial planes of HCCs from the log map to increase the dataset for training. Subsequently, 2D CNN was used to extract deep features of the log map for HCCs. Finally, fusion of deep features derived from three b-value log maps was conducted for HCC malignancy classification. Specifically, a deeply supervised loss function was devised to further improve the performance of lesion characterization. The data set was split into two parts: the training and validation set (60 HCCs) and the fixed test set (40 HCCs). Four-fold cross validation with 10 repetitions was performed to assess the performance of deep features extracted from single b-value images for HCC grading using the training and validation set. Receiver operating characteristic curve (ROC) and area under the curve (AUC) values were used to assess the characterization performance of the proposed deep feature fusion method to differentiate low-grade and high-grade in the fixed test set.

RESULTS

The proposed fusion of deep features derived from logb0, logb100, and logb600 with deeply supervised loss function generated the highest accuracy for HCC grading (80%), thus outperforming the method of deep feature derived from the ADC map directly (72.5%), the original b0 (65%), b100 (68%), and b600 (70%) images. Furthermore, AUC values of the deep features of the ADC map, the deep feature fusion with concatenation, and the proposed deep feature fusion with deeply supervised loss function were 0.73, 0.78, and 0.83, respectively.

CONCLUSION

The proposed fusion of deep features derived from the logarithm of the three b-value images yields high performance for HCC grading, thus providing a promising approach for the assessment of DWI in lesion characterization.

Depression recognition according to heart rate variability using Bayesian Networks

BACKGROUND

Doctors mainly use scale tests and subjective judgment in the clinical diagnosis of depression. Researches have demonstrated that depression is associated with the dysfunction of the autonomic nervous system (ANS), where its modulation can be evaluated by heart rate variability (HRV). Depression patients have lower HRV than healthy subjects. Therefore, HRV may be used to distinguish depression patients from healthy people.

METHODS

HRV signals were collected from 76 female subjects composed of 38 depression patients and 38 healthy people. Time domain, frequency domain, and non-linear features were extracted from the HRV signals of these subjects, who were subjected to the Ewing test as an ANS stimulus. Then, these multiple features were input into Bayesian networks, served as a classifier, to distinguish depression patients from healthy people. Hence, accuracy, sensitivity, and specificity were calculated to evaluate the performance of the classifier.

RESULTS

Recognition results indicate 86.4% accuracy, 89.5% sensitivity, and 84.2% specificity. The individuals subjected to the Ewing test showed better recognition results than those at individual test states (resting state, deep breathing state, Valsalva state, and standing state) of the Ewing test. The root mean square of successive differences (RMSSD) of the HRV exhibits a significant relevance with recognition.

CONCLUSION

Bayesian networks can be applied to the recognition of depression patients from healthy people and the recognition results demonstrate the significant association between depression and HRV. The Ewing test is a good ANS stimulus for acquiring the difference of HRV between depression patients and healthy people to recognize depression. The RMSSD of the HRV is important in recognition and may be a significant index in distinguishing depression patients from healthy people.

Gastrointestinal polyp detection in endoscopic images using an improved feature extraction method

Gastrointestinal polyps are treated as the precursors of cancer development. So, possibility of cancers can be reduced at a great extent by early detection and removal of polyps. The most used diagnostic modality for gastrointestinal polyps is video endoscopy. But, as an operator dependant procedure, several human factors can lead to miss detection of polyps. In this peper, an improved computer aided polyp detection method has been proposed. Proposed improved method can reduce polyp miss detection rate and assists doctors in finding the most important regions to pay attention. Color wavelet features and convolutional neural network features are extracted from endoscopic images, which are used for training a support vector machine. Then a target endoscopic image will be given to the classifier as input in order to find whether it contains any polyp or not. If polyp is found, it will be marked automatically. Experiment shows that, color wavelet features and convolutional neural network features together construct a highly representative of endoscopic polyp images. Evaluations on standard public databases show that, proposed system outperforms state-of-the-art methods, gaining accuracy of 98.34%, sensitivity of 98.67% and specificity of 98.23%. In this paper, the strength of color wavelet features and power of convolutional neural network features are combined. Fusion of these two methodology and use of support vector machine results in an improved method for gastrointestinal polyp detection. An analysis of ROC reveals that, proposed method can be used for polyp detection purposes with greater accuracy than state-of-the-art methods.

An Automatic Gastrointestinal Polyp Detection System in Video Endoscopy Using Fusion of Color Wavelet and Convolutional Neural Network Features

Gastrointestinal polyps are considered to be the precursors of cancer development in most of the cases. Therefore, early detection and removal of polyps can reduce the possibility of cancer. Video endoscopy is the most used diagnostic modality for gastrointestinal polyps. But, because it is an operator dependent procedure, several human factors can lead to misdetection of polyps. Computer aided polyp detection can reduce polyp miss detection rate and assists doctors in finding the most important regions to pay attention to. In this paper, an automatic system has been proposed as a support to gastrointestinal polyp detection. This system captures the video streams from endoscopic video and, in the output, it shows the identified polyps. Color wavelet (CW) features and convolutional neural network (CNN) features of video frames are extracted and combined together which are used to train a linear support vector machine (SVM). Evaluations on standard public databases show that the proposed system outperforms the state-of-the-art methods, gaining accuracy of 98.65%, sensitivity of 98.79%, and specificity of 98.52%.

Feasibility Study of a Generalized Framework for Developing Computer-Aided Detection Systems-a New Paradigm

We propose a generalized framework for developing computer-aided detection (CADe) systems whose characteristics depend only on those of the training dataset. The purpose of this study is to show the feasibility of the framework. Two different CADe systems were experimentally developed by a prototype of the framework, but with different training datasets. The CADe systems include four components; preprocessing, candidate area extraction, candidate detection, and candidate classification. Four pretrained algorithms with dedicated optimization/setting methods corresponding to the respective components were prepared in advance. The pretrained algorithms were sequentially trained in the order of processing of the components. In this study, two different datasets, brain MRA with cerebral aneurysms and chest CT with lung nodules, were collected to develop two different types of CADe systems in the framework. The performances of the developed CADe systems were evaluated by threefold cross-validation. The CADe systems for detecting cerebral aneurysms in brain MRAs and for detecting lung nodules in chest CTs were successfully developed using the respective datasets. The framework was shown to be feasible by the successful development of the two different types of CADe systems. The feasibility of this framework shows promise for a new paradigm in the development of CADe systems: development of CADe systems without any lesion specific algorithm designing.

A CAD of fully automated colonic polyp detection for contrasted and non-contrasted CT scans

PURPOSE

Computer-aided detection (CAD) systems are developed to help radiologists detect colonic polyps over CT scans. It is possible to reduce the detection time and increase the detection accuracy rates by using CAD systems. In this paper, we aimed to develop a fully integrated CAD system for automated detection of polyps that yields a high polyp detection rate with a reasonable number of false positives.

METHODS

The proposed CAD system is a multistage implementation whose main components are: automatic colon segmentation, candidate detection, feature extraction and classification. The first element of the algorithm includes a discrete segmentation for both air and fluid regions. Colon-air regions were determined based on adaptive thresholding, and the volume/length measure was used to detect air regions. To extract the colon-fluid regions, a rule-based connectivity test was used to detect the regions belong to the colon. Potential polyp candidates were detected based on the 3D Laplacian of Gaussian filter. The geometrical features were used to reduce false-positive detections. A 2D projection image was generated to extract discriminative features as the inputs of an artificial neural network classifier.

RESULTS

Our CAD system performs at 100% sensitivity for polyps larger than 9 mm, 95.83% sensitivity for polyps 6-10 mm and 85.71% sensitivity for polyps smaller than 6 mm with 5.3 false positives per dataset. Also, clinically relevant polyps ([Formula: see text]6 mm) were identified with 96.67% sensitivity at 1.12 FP/dataset.

CONCLUSIONS

To the best of our knowledge, the novel polyp candidate detection system which determines polyp candidates with LoG filters is one of the main contributions. We also propose a new 2D projection image calculation scheme to determine the distinctive features. We believe that our CAD system is highly effective for assisting radiologist interpreting CT.

Improving Computer-Aided Detection Using Convolutional Neural Networks and Random View Aggregation

Automated computer-aided detection (CADe) has been an important tool in clinical practice and research. State-of-the-art methods often show high sensitivities at the cost of high false-positives (FP) per patient rates. We design a two-tiered coarse-to-fine cascade framework that first operates a candidate generation system at sensitivities  ∼ 100% of but at high FP levels. By leveraging existing CADe systems, coordinates of regions or volumes of interest (ROI or VOI) are generated and function as input for a second tier, which is our focus in this study. In this second stage, we generate 2D (two-dimensional) or 2.5D views via sampling through scale transformations, random translations and rotations. These random views are used to train deep convolutional neural network (ConvNet) classifiers. In testing, the ConvNets assign class (e.g., lesion, pathology) probabilities for a new set of random views that are then averaged to compute a final per-candidate classification probability. This second tier behaves as a highly selective process to reject difficult false positives while preserving high sensitivities. The methods are evaluated on three data sets: 59 patients for sclerotic metastasis detection, 176 patients for lymph node detection, and 1,186 patients for colonic polyp detection. Experimental results show the ability of ConvNets to generalize well to different medical imaging CADe applications and scale elegantly to various data sets. Our proposed methods improve performance markedly in all cases. Sensitivities improved from 57% to 70%, 43% to 77%, and 58% to 75% at 3 FPs per patient for sclerotic metastases, lymph nodes and colonic polyps, respectively.

Progress in Fully Automated Abdominal CT Interpretation

OBJECTIVE

Automated analysis of abdominal CT has advanced markedly over just the last few years. Fully automated assessment of organs, lymph nodes, adipose tissue, muscle, bowel, spine, and tumors are some examples where tremendous progress has been made. Computer-aided detection of lesions has also improved dramatically.

CONCLUSION

This article reviews the progress and provides insights into what is in store in the near future for automated analysis for abdominal CT, ultimately leading to fully automated interpretation.

Computer-Aided Detection of Polyps in Optical Colonoscopy Images

We present a computer-aided detection algorithm for polyps in optical colonoscopy images. Polyps are the precursors to colon cancer. In the US alone, more than 14 million optical colonoscopies are performed every year, mostly to screen for polyps. Optical colonoscopy has been shown to have an approximately 25% polyp miss rate due to the convoluted folds and bends present in the colon. In this work, we present an automatic detection algorithm to detect these polyps in the optical colonoscopy images. We use a machine learning algorithm to infer a depth map for a given optical colonoscopy image and then use a detailed pre-built polyp profile to detect and delineate the boundaries of polyps in this given image. We have achieved the best recall of 84.0% and the best specificity value of 83.4%.

Structured learning algorithm for detection of nonobstructive and obstructive coronary plaque lesions from computed tomography angiography

Visual identification of coronary arterial lesion from three-dimensional coronary computed tomography angiography (CTA) remains challenging. We aimed to develop a robust automated algorithm for computer detection of coronary artery lesions by machine learning techniques. A structured learning technique is proposed to detect all coronary arterial lesions with stenosis [Formula: see text]. Our algorithm consists of two stages: (1) two independent base decisions indicating the existence of lesions in each arterial segment and (b) the final decision made by combining the base decisions. One of the base decisions is the support vector machine (SVM) based learning algorithm, which divides each artery into small volume patches and integrates several quantitative geometric and shape features for arterial lesions in each small volume patch by SVM algorithm. The other base decision is the formula-based analytic method. The final decision in the first stage applies SVM-based decision fusion to combine the two base decisions in the second stage. The proposed algorithm was applied to 42 CTA patient datasets, acquired with dual-source CT, where 21 datasets had 45 lesions with stenosis [Formula: see text]. Visual identification of lesions with stenosis [Formula: see text] by three expert readers, using consensus reading, was considered as a reference standard. Our method performed with high sensitivity (93%), specificity (95%), and accuracy (94%), with receiver operator characteristic area under the curve of 0.94. The proposed algorithm shows promising results in the automated detection of obstructive and nonobstructive lesions from CTA.

Quantitative imaging: quantification of liver shape on CT using the statistical shape model to evaluate hepatic fibrosis

RATIONALE AND OBJECTIVES

To investigate the usefulness of the statistical shape model (SSM) for the quantification of liver shape to evaluate hepatic fibrosis.

MATERIALS AND METHODS

Ninety-one subjects (45 men and 46 women; age range, 20-75 years) were included in this retrospective study: 54 potential liver donors and 37 patients with chronic liver disease. The subjects were classified histopathologically according to the fibrosis stage as follows: F0 (n = 55); F1 (n = 6); F2 (3); F3 (n = 1); and F4 (n = 26). Each subject underwent contrast-enhanced computed tomography (CT) using a 64-channel scanner (0.625-mm slice thickness). An abdominal radiologist manually traced the liver boundaries on every CT section using an image workstation; the boundaries were used for subsequent analyses. An SSM was constructed by the principal component analysis of the subject data set, which defined a parametric model of the liver shapes. The shape parameters were calculated by fitting SSM to the segmented liver shape of each subject and were used for the training of a linear support vector regression (SVR), which classifies the liver fibrosis stage to maximize the area under the receiver operating characteristic curve (AUC). SSM/SVR models were constructed and were validated in a leave-one-out manner. The performance of our technique was compared to those of two previously reported types of caudate-right lobe ratios (C/RL-m and C/RL-r).

RESULTS

In our SSM/SVR models, the AUC values for the classification of liver fibrosis were 0.96 (F0 vs. F1-4), 0.95 (F0-1 vs. F2-4), 0.96 (F0-2 vs. F3-4), and 0.95 (F0-3 vs. F4). These values were significantly superior to AUC values using the C/RL-m or C/RL-r ratios (P < .005).

CONCLUSIONS

SSM was useful for estimating the stage of hepatic fibrosis by quantifying liver shape.

A new 2.5D representation for lymph node detection using random sets of deep convolutional neural network observations

Automated Lymph Node (LN) detection is an important clinical diagnostic task but very challenging due to the low contrast of surrounding structures in Computed Tomography (CT) and to their varying sizes, poses, shapes and sparsely distributed locations. State-of-the-art studies show the performance range of 52.9% sensitivity at 3.1 false-positives per volume (FP/vol.), or 60.9% at 6.1 FP/vol. for mediastinal LN, by one-shot boosting on 3D HAAR features. In this paper, we first operate a preliminary candidate generation stage, towards -100% sensitivity at the cost of high FP levels (-40 per patient), to harvest volumes of interest (VOI). Our 2.5D approach consequently decomposes any 3D VOI by resampling 2D reformatted orthogonal views N times, via scale, random translations, and rotations with respect to the VOI centroid coordinates. These random views are then used to train a deep Convolutional Neural Network (CNN) classifier. In testing, the CNN is employed to assign LN probabilities for all N random views that can be simply averaged (as a set) to compute the final classification probability per VOI. We validate the approach on two datasets: 90 CT volumes with 388 mediastinal LNs and 86 patients with 595 abdominal LNs. We achieve sensitivities of 70%/83% at 3 FP/vol. and 84%/90% at 6 FP/vol. in mediastinum and abdomen respectively, which drastically improves over the previous state-of-the-art work.

3D Reconstruction of virtual colon structures from colonoscopy images

This paper presents the first fully automated reconstruction technique of 3D virtual colon segments from individual colonoscopy images. It is the basis of new software applications that may offer great benefits for improving quality of care for colonoscopy patients. For example, a 3D map of the areas inspected and uninspected during colonoscopy can be shown on request of the endoscopist during the procedure. The endoscopist may revisit the suggested uninspected areas to reduce the chance of missing polyps that reside in these areas. The percentage of the colon surface seen by the endoscopist can be used as a coarse objective indicator of the quality of the procedure. The derived virtual colon models can be stored for post-procedure training of new endoscopists to teach navigation techniques that result in a higher level of procedure quality. Our technique does not require a prior CT scan of the colon or any global positioning device. Our experiments on endoscopy images of an Olympus synthetic colon model reveal encouraging results with small average reconstruction errors (4.1 mm for the fold depths and 12.1 mm for the fold circumferences).

Automated knowledge-based detection of nonobstructive and obstructive arterial lesions from coronary CT angiography

PURPOSE

Visual analysis of three-dimensional (3D) coronary computed tomography angiography (CCTA) remains challenging due to large number of image slices and tortuous character of the vessels. The authors aimed to develop a robust, automated algorithm for unsupervised computer detection of coronary artery lesions.

METHODS

The authors' knowledge-based algorithm consists of centerline extraction, vessel classification, vessel linearization, lumen segmentation with scan-specific lumen attenuation ranges, and lesion location detection. Presence and location of lesions are identified using a multi-pass algorithm which considers expected or "normal" vessel tapering and luminal stenosis from the segmented vessel. Expected luminal diameter is derived from the scan by automated piecewise least squares line fitting over proximal and mid segments (67%) of the coronary artery considering the locations of the small branches attached to the main coronary arteries.

RESULTS

The authors applied this algorithm to 42 CCTA patient datasets, acquired with dual-source CT, where 21 datasets had 45 lesions with stenosis ≥ 25%. The reference standard was provided by visual and quantitative identification of lesions with any stenosis ≥ 25% by three expert readers using consensus reading. The authors algorithm identified 42 lesions (93%) confirmed by the expert readers. There were 46 additional lesions detected; 23 out of 39 (59%) of these were less-stenosed lesions. When the artery was divided into 15 coronary segments according to standard cardiology reporting guidelines, per-segment basis, sensitivity was 93% and per-segment specificity was 81% using 10-fold cross-validation.

CONCLUSIONS

The authors' algorithm shows promising results in the detection of both obstructive and nonobstructive CCTA lesions.

Machine Learning in Computer-aided Diagnosis of the Thorax and Colon in CT: A Survey

Computer-aided detection (CADe) and diagnosis (CAD) has been a rapidly growing, active area of research in medical imaging. Machine leaning (ML) plays an essential role in CAD, because objects such as lesions and organs may not be represented accurately by a simple equation; thus, medical pattern recognition essentially require "learning from examples." One of the most popular uses of ML is the classification of objects such as lesion candidates into certain classes (e.g., abnormal or normal, and lesions or non-lesions) based on input features (e.g., contrast and area) obtained from segmented lesion candidates. The task of ML is to determine "optimal" boundaries for separating classes in the multidimensional feature space which is formed by the input features. ML algorithms for classification include linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), multilayer perceptrons, and support vector machines (SVM). Recently, pixel/voxel-based ML (PML) emerged in medical image processing/analysis, which uses pixel/voxel values in images directly, instead of features calculated from segmented lesions, as input information; thus, feature calculation or segmentation is not required. In this paper, ML techniques used in CAD schemes for detection and diagnosis of lung nodules in thoracic CT and for detection of polyps in CT colonography (CTC) are surveyed and reviewed.

A review of computer-aided diagnosis in thoracic and colonic imaging

Medical imaging has been indispensable in medicine since the discovery of x-rays. Medical imaging offers useful information on patients' medical conditions and on the causes of their symptoms and diseases. As imaging technologies advance, a large number of medical images are produced which physicians/radiologists must interpret. Thus, computer aids are demanded and become indispensable in physicians' decision making based on medical images. Consequently, computer-aided detection and diagnosis (CAD) has been investigated and has been an active research area in medical imaging. CAD is defined as detection and/or diagnosis made by a radiologist/physician who takes into account the computer output as a "second opinion". In CAD research, detection and diagnosis of lung and colorectal cancer in thoracic and colonic imaging constitute major areas, because lung and colorectal cancers are the leading and second leading causes, respectively, of cancer deaths in the U.S. and also in other countries. In this review, CAD of the thorax and colon, including CAD for detection and diagnosis of lung nodules in thoracic CT, and that for detection of polyps in CT colonography, are reviewed.

High performance lung nodule detection schemes in CT using local and global information

PURPOSE

A key issue in current computer-aided diagnostic (CAD) schemes for nodule detection in CT is the large number of false positives, because current schemes use only global three-dimensional (3D) information to detect nodules and discard useful local two-dimensional (2D) information. Thus, the authors integrated local and global information to markedly improve the performance levels of CAD schemes.

METHODS

Our database was obtained from the standard CT lung nodule database created by the Lung Image Database Consortium (LIDC). It consisted of 85 CT scans with 111 nodules of 3 mm or larger in diameter. The 111 nodules were confirmed by at least two of the four radiologists participated in the LIDC. Twenty-six nodules were missed by two of the four radiologists and were thus very difficult to detect. The authors developed five CAD schemes for nodule detection in CT using global 3D information (3D scheme), local 2D information (2D scheme), and both local and global information (2D + 3D scheme, 2D - 3D scheme, and 3D - 2D scheme). The 3D scheme, which was developed previously, used only global 3D information and discarded local 2D information, as other CAD schemes did. The 2D scheme used a uniform viewpoint reformation technique to decompose a 3D nodule candidate into a set of 2D reformatted images generated from representative viewpoints, and selected and used "effective" 2D reformatted images to remove false positives. The 2D + 3D scheme, 2D - 3D scheme, and 3D - 2D scheme used complementary local and global information in different ways to further improve the performance of lung nodule detection. The authors employed a leave-one-scan-out testing method for evaluation of the performance levels of the five CAD schemes.

RESULTS

At the sensitivities of 85%, 80%, and 75%, the existing 3D scheme reported 17.3, 7.4, and 2.8 false positives per scan, respectively; the 2D scheme improved the detection performance and reduced the numbers of false positives to 7.6, 2.5, and 1.3 per scan; the 2D + 3D scheme further reduced those to 2.7, 1.9, and 0.6 per scan; the 2D - 3D scheme reduced those to 7.6, 2.1, and 0.8 per scan; and the 3D - 2D scheme reduced those to 17.3, 1.6, and 1.0 per scan.

CONCLUSIONS

The local 2D information appears to be more useful than the global 3D information for nodule detection, particularly, when it is integrated with 3D information.

Machine Learning for Automated Polyp Detection in Computed Tomography Colonography

This chapter presents a comprehensive scheme for automated detection of colorectal polyps in computed tomography colonography (CTC) with particular emphasis on robust learning algorithms that differentiate polyps from non-polyp shapes. The authors’ automated CTC scheme introduces two orientation independent features which encode the shape characteristics that aid in classification of polyps and non-polyps with high accuracy, low false positive rate, and low computations making the scheme suitable for colorectal cancer screening initiatives. Experiments using state-of-the-art machine learning algorithms viz., lazy learning, support vector machines, and naïve Bayes classifiers reveal the robustness of the two features in detecting polyps at 100% sensitivity for polyps with diameter greater than 10 mm while attaining total low false positive rates, respectively, of 3.05, 3.47 and 0.71 per CTC dataset at specificities above 99% when tested on 58 CTC datasets. The results were validated using colonoscopy reports provided by expert radiologists.

Improved computer-aided detection of small polyps in CT colonography using interpolation for curvature estimation

PURPOSE

Surface curvatures are important geometric features for the computer-aided analysis and detection of polyps in CT colonography (CTC). However, the general kernel approach for curvature computation can yield erroneous results for small polyps and for polyps that lie on haustral folds. Those erroneous curvatures will reduce the performance of polyp detection. This paper presents an analysis of interpolation's effect on curvature estimation for thin structures and its application on computer-aided detection of small polyps in CTC.

METHODS

The authors demonstrated that a simple technique, image interpolation, can improve the accuracy of curvature estimation for thin structures and thus significantly improve the sensitivity of small polyp detection in CTC.

RESULTS

Our experiments showed that the merits of interpolating included more accurate curvature values for simulated data, and isolation of polyps near folds for clinical data. After testing on a large clinical data set, it was observed that sensitivities with linear, quadratic B-spline and cubic B-spline interpolations significantly improved the sensitivity for small polyp detection.

CONCLUSIONS

The image interpolation can improve the accuracy of curvature estimation for thin structures and thus improve the computer-aided detection of small polyps in CTC.

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

PURPOSE

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.

METHODS

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.

RESULTS

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.

CONCLUSIONS

Both MTSVR and MTGPR improve the efficiency of the training in the massive-training framework while maintaining a comparable performance.

Image processing and classification algorithm for yeast cell morphology in a microfluidic chip

The study of yeast cell morphology requires consistent identification of cell cycle phases based on cell bud size. A computer-based image processing algorithm is designed to automatically classify microscopic images of yeast cells in a microfluidic channel environment. The images were enhanced to reduce background noise, and a robust segmentation algorithm is developed to extract geometrical features including compactness, axis ratio, and bud size. The features are then used for classification, and the accuracy of various machine-learning classifiers is compared. The linear support vector machine, distance-based classification, and k-nearest-neighbor algorithm were the classifiers used in this experiment. The performance of the system under various illumination and focusing conditions were also tested. The results suggest it is possible to automatically classify yeast cells based on their morphological characteristics with noisy and low-contrast images.

Automatic colon polyp flagging via geometric and texture features

Computer Tomographic Colonography, combined with computer-aided detection (CAD), is a promising emerging technique for colonic polyp analysis. We present a CAD scheme for polyp flagging based on new texture and geometric features that consider both the information in the candidate polyp location and its immediate surrounding area, testing multiple sizes. The proposed algorithm is tested with ground truth data, including flat and small polyps, with very promising results.

Content-based image retrieval in radiology: current status and future directions

Diagnostic radiology requires accurate interpretation of complex signals in medical images. Content-based image retrieval (CBIR) techniques could be valuable to radiologists in assessing medical images by identifying similar images in large archives that could assist with decision support. Many advances have occurred in CBIR, and a variety of systems have appeared in nonmedical domains; however, permeation of these methods into radiology has been limited. Our goal in this review is to survey CBIR methods and systems from the perspective of application to radiology and to identify approaches developed in nonmedical applications that could be translated to radiology. Radiology images pose specific challenges compared with images in the consumer domain; they contain varied, rich, and often subtle features that need to be recognized in assessing image similarity. Radiology images also provide rich opportunities for CBIR: rich metadata about image semantics are provided by radiologists, and this information is not yet being used to its fullest advantage in CBIR systems. By integrating pixel-based and metadata-based image feature analysis, substantial advances of CBIR in medicine could ensue, with CBIR systems becoming an important tool in radiology practice.

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).

Image analysis for classification of dysplasia in Barrett's esophagus using endoscopic optical coherence tomography

Barrett's esophagus (BE) and associated adenocarcinoma have emerged as a major health care problem. Endoscopic optical coherence tomography is a microscopic sub-surface imaging technology that has been shown to differentiate tissue layers of the gastrointestinal wall and identify dysplasia in the mucosa, and is proposed as a surveillance tool to aid in management of BE. In this work a computer-aided diagnosis (CAD) system has been demonstrated for classification of dysplasia in Barrett's esophagus using EOCT. The system is composed of four modules: region of interest segmentation, dysplasia-related image feature extraction, feature selection, and site classification and validation. Multiple feature extraction and classification methods were evaluated and the process of developing the CAD system is described in detail. Use of multiple EOCT images to classify a single site was also investigated. A total of 96 EOCT image-biopsy pairs (63 non-dysplastic, 26 low-grade and 7 high-grade dysplastic biopsy sites) from a previously described clinical study were analyzed using the CAD system, yielding an accuracy of 84% for classification of non-dysplastic vs. dysplastic BE tissue. The results motivate continued development of CAD to potentially enable EOCT surveillance of large surface areas of Barrett's mucosa to identify dysplasia.

Increasing computer-aided detection specificity by projection features for CT colonography

PURPOSE

A large number of false positives (FPs) generated by computer-aided detection (CAD) schemes is likely to distract radiologists' attention and decrease their interpretation efficiency. This study aims to develop projection-based features which characterize true and false positives to increase the specificity while maintaining high sensitivity in detecting colonic polyps.

METHODS

In this study, two-dimensional projection images are obtained from each initial polyp candidate or volume of interest, and features are extracted from both the gray and color projection images to differentiate FPs from true positives. These projection features were tested to exclude different types of FPs, such as haustral folds, rectal tubes, and residue stool using a database of 325 patient studies (from two different institutions), which includes 556 scans at supine and/or prone positions with 347 polyps and masses sized from 5 to 60 mm. For comparison, several well-established features were used to generate a baseline reference. The experimental evaluation was conducted for large polyps (> or = 10 mm) and medium-sized polyps (5-9 mm) separately.

RESULTS

For large polyps, the additional usage of the projection features reduces the FP rate from 5.31 to 1.92 per scan at the comparable by-polyp sensitivity level of 93.1%. For medium-sized polyps, the FP rate is reduced from 8.89 to 5.23 at the sensitivity level of 80.6%. The percentages of FP reduction are 63.9% and 41.2% for the large and medium-sized polyps, respectively, without sacrificing detection sensitivity.

CONCLUSIONS

The results have demonstrated that the new projection features can effectively reduce the FPs and increase the detection specificity without sacrificing the sensitivity. CAD of colonic polyps is supposed to help radiologists to improve their performance in interpreting computed tomographic colonography images.

Computer-aided detection of polyps in CT colonography using logistic regression

We present a computer-aided detection (CAD) system for computed tomography colonography that orders the polyps according to clinical relevance. The CAD system consists of two steps: candidate detection and supervised classification. The characteristics of the detection step lead to specific choices for the classification system. The candidates are ordered by a linear logistic classifier (logistic regression) based on only three features: the protrusion of the colon wall, the mean internal intensity, and a feature to discard detections on the rectal enema tube. This classifier can cope with a small number of polyps available for training, a large imbalance between polyps and non-polyp candidates, a truncated feature space, unbalanced and unknown misclassification costs, and an exponential distribution with respect to candidate size in feature space. Our CAD system was evaluated with data sets from four different medical centers. For polyps larger than or equal to 6 mm we achieved sensitivities of respectively 95%, 85%, 85%, and 100% with 5, 4, 5, and 6 false positives per scan over 86, 48, 141, and 32 patients. A cross-center evaluation in which the system is trained and tested with data from different sources showed that the trained CAD system generalizes to data from different medical centers and with different patient preparations. This is essential to application in large-scale screening for colorectal polyps.

Linear measurement of polyps in CT colonography using level sets on 3D surfaces

CT colonography has emerged as a minimally invasive alternative to optical colonoscopy for the screening of polyps which are the precursors to colon cancer. Accurate polyp measurement is crucial as the size of a polyp is considered an indication of its potential for malignancy. We present a novel method for the automatic measurement of polyps. It is based on a level set algorithm capable of evolving on the surface of a 3D object represented by a triangular mesh. It is guided by curvature features and is capable of segmenting the polyp neck, that is, the ridgeline/crestline formed around the polyp by its merging to the colon wall. Our method was validated on 40 polyp surfaces obtained from real clinical data. A 3D manual measurement was used as the reference standard. A correlation of 0.825 was found between polyp measurements from our new method and the reference standard.

Automated detection and segmentation of large lesions in CT colonography

Computerized tomographic colonography is a minimally invasive technique for the detection of colorectal polyps and carcinoma. Computer-aided diagnosis (CAD) schemes are designed to help radiologists locating colorectal lesions in an efficient and accurate manner. Large lesions are often initially detected as multiple small objects, due to which such lesions may be missed or misclassified by CAD systems. We propose a novel method for automated detection and segmentation of all large lesions, i.e., large polyps as well as carcinoma. Our detection algorithm is incorporated in a classical CAD system. Candidate detection comprises preselection based on a local measure for protrusion and clustering based on geodesic distance. The generated clusters are further segmented and analyzed. The segmentation algorithm is a thresholding operation in which the threshold is adaptively selected. The segmentation provides a size measurement that is used to compute the likelihood of a cluster to be a large lesion. The large lesion detection algorithm was evaluated on data from 35 patients having 41 large lesions (19 of which malignant) confirmed by optical colonoscopy. At five false positive (FP) per scan, the classical system achieved a sensitivity of 78%, while the system augmented with the large lesion detector achieved 83% sensitivity. For malignant lesions, the performance at five FP/scan was increased from 79% to 95%. The good results on malignant lesions demonstrate that the proposed algorithm may provide relevant additional information for the clinical decision process.

EMPLOYING TOPOGRAPHICAL HEIGHT MAP IN COLONIC POLYP MEASUREMENT AND FALSE POSITIVE REDUCTION

CT Colonography (CTC) is an emerging minimally invasive technique for screening and diagnosing colon cancers. Computer Aided Detection (CAD) techniques can increase sensitivity and reduce false positives. Inspired by the way radiologists detect polyps via 3D virtual fly-through in CTC, we borrowed the idea from geographic information systems to employ topographical height map in colonic polyp measurement and false positive reduction. After a curvature based filtering and a 3D CT feature classifier, a height map is computed for each detection using a ray-casting algorithm. We design a concentric index to characterize the concentric pattern in polyp height map based on the fact that polyps are protrusions from the colon wall and round in shape. The height map is optimized through a multi-scale spiral spherical search to maximize the concentric index. We derive several topographic features from the map and compute texture features based on wavelet decomposition. We then send the features to a committee of support vector machines for classification. We have trained our method on 394 patients (71 polyps) and tested it on 792 patients (226 polyps). Results showed that we can achieve 95% sensitivity at 2.4 false positives per patient and the height map features can reduce false positives by more than 50%. We compute the polyp height and width measurements and correlate them with manual measurements. The Pearson correlations are 0.74 (p=0.11) and 0.75 (p=0.17) for height and width, respectively.

Wavelet method for CT colonography computer-aided polyp detection

Computed tomographic colonography (CTC) computer aided detection (CAD) is a new method to detect colon polyps. Colonic polyps are abnormal growths that may become cancerous. Detection and removal of colonic polyps, particularly larger ones, has been shown to reduce the incidence of colorectal cancer. While high sensitivities and low false positive rates are consistently achieved for the detection of polyps sized 1 cm or larger, lower sensitivities and higher false positive rates occur when the goal of CAD is to identify "medium"-sized polyps, 6-9 mm in diameter. Such medium-sized polyps may be important for clinical patient management. We have developed a wavelet-based postprocessor to reduce false positives for this polyp size range. We applied the wavelet-based postprocessor to CTC CAD findings from 44 patients in whom 45 polyps with sizes of 6-9 mm were found at segmentally unblinded optical colonoscopy and visible on retrospective review of the CT colonography images. Prior to the application of the wavelet-based postprocessor, the CTC CAD system detected 33 of the polyps (sensitivity 73.33%) with 12.4 false positives per patient, a sensitivity comparable to that of expert radiologists. Fourfold cross validation with 5000 bootstraps showed that the wavelet-based postprocessor could reduce the false positives by 56.61% (p <0.001), to 5.38 per patient (95% confidence interval [4.41, 6.34]), without significant sensitivity degradation (32/45, 71.11%, 95% confidence interval [66.39%, 75.74%], p=0.1713). We conclude that this wavelet-based postprocessor can substantially reduce the false positive rate of our CTC CAD for this important polyp size range.

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.

Computer-aided detection for CT colonography: update 2007

Computed tomographic colonography (CTC) is an emerging technique for polyp detection in the colon. However, lesion detection can be challenging due to insufficient patient preparation, chosen CT technique or reader imperfection. The primary goal of computer-aided detection (CAD) for CTC is locating possible polyps, and presenting the reader with these polyp candidates. Other goals are sensitivity improvement and reduction of reading time and inter-observer variability. The multistep CAD procedure typically consists of segmentation of the colonic wall (e.g. region growing); selection of intermediate polyp candidates (curvature analysis, sphere fitting, normal analysis, slope density function ...); classification of final candidates for detection and listing suspicious polyps (location, size and volume). Remaining task for the radiologist is the validation or rejection of the polyp candidates. State-of-the-art CAD systems should require minimal or even no user interaction for the extraction of the colonic wall, offer a computation time less than 10-20 min and high sensitivity and specificity for different polyp sizes and shapes, with a low number of false positives. These systems have the potential to increase radiologist's performance and to decrease inter-reader variability. Besides CAD key techniques we also discuss new developments in CAD and describe recent applications facilitating CTC.

A probabilistic model for haustral curvatures with applications to colon CAD

Among the many features used for classification in computer-aided detection (CAD) systems targeting colonic polyps, those based on differences between the shapes of polyps and folds are most common. We introduce here an explicit parametric model for the haustra or colon wall. The proposed model captures the overall shape of the haustra and we use it to derive the probability distribution of features relevant to polyp detection. The usefulness of the model is demonstrated through its application to a colon CAD algorithm.

Polyp enhancing level set evolution of colon wall: method and pilot study

Computer aided detection (CAD) in computed tomography colonography (CTC) aims at detecting colonic polyps that are the precursors of colon cancer. In this work, we propose a colon wall evolution algorithm polyp enhancing level sets (PELS) based on the level-set formulation that regularizes and enhances polyps as a preprocessing step to CTC CAD algorithms. The underlying idea is to evolve the polyps towards spherical protrusions on the colon wall while keeping other structures, such as haustral folds, relatively unchanged and, thereby, potentially improve the performance of CTC CAD algorithms, especially for smaller polyps. To evaluate our methods, we conducted a pilot study using an arbitrarily chosen CTC CAD method, the surface normal overlap (SNO) CAD algorithm, on a nine patient CTC data set with 47 polyps of sizes ranging from 2.0 to 17.0 mm in diameter. PELS increased the maximum sensitivity by 8.1% (from 21/37 to 24/37) for small polyps of sizes ranging from 5.0 to 9.0 mm in diameter. This is accompanied by a statistically significant separation between small polyps and false positives. PELS did not change the CTC CAD performance significantly for larger polyps.