Classification Of Colonic Polyps Research Articles

Paris classification of colonic polyps using CT colonography: prospective cohort study of interobserver variation.

The Paris classification categorises colorectal polyp morphology. Interobserver agreement for Paris classification has been assessed at optical colonoscopy (OC) but not CT colonography (CTC). We aimed to determine the following: (1) interobserver agreement for the Paris classification using CTC between radiologists; (2) if radiologist experience influenced classification, gross polyp morphology, or polyp size; and (3) the extent to which radiologist classifications agreed with (a) colonoscopy and (b) a combined reference standard. Following ethical approval for this non-randomised prospective cohort study, seven radiologists from three hospitals classified 52 colonic polyps using the Paris system. We calculated interobserver agreement using Fleiss kappa and mean pairwise agreement (MPA). Absolute agreement was calculated between radiologists; between CTC and OC; and between CTC and a combined reference standard using all available imaging, colonoscopic, and histopathological data. Overall interobserver agreement between the seven readers was fair (Fleiss kappa 0.33; 95% CI 0.30-0.37; MPA 49.7%). Readers with < 1500 CTC experience had higher interobserver agreement (0.42 (95% CI 0.35-0.48) vs. 0.33 (95% CI 0.25-0.42)) and MPA (69.2% vs 50.6%) than readers with ≥ 1500 experience. There was substantial overall agreement for flat vs protuberant polyps (0.62 (95% CI 0.56-0.68)) with a MPA of 87.9%. Agreement between CTC and OC classifications was only 44%, and CTC agreement with the combined reference standard was 56%. Radiologist agreement when using the Paris classification at CT colonography is low, and radiologist classification agrees poorly with colonoscopy. Using the full Paris classification in routine CTC reporting is of questionable value. Interobserver agreement for radiologists using the Paris classification to categorise colorectal polyp morphology is only fair; routine use of the full Paris classification at CT colonography is questionable. • Overall interobserver agreement for the Paris classification at CT colonography (CTC) was only fair, and lower than for colonoscopy. • Agreement was higher for radiologists with < 1500 CTC experience and for larger polyps. There was substantial agreement when classifying polyps as protuberant vs flat. • Agreement between CTC and colonoscopic polyp classification was low (44%).

Recognition of Colon Polyps (Tubular Adenoma, Villous Adenoma) and Normal Colon Epithelium Histomorphology with Transfer Learning.

The use of artificial intelligence technology in medicine, which is remarkable with its increasing use in many areas recently, has allowed for rapid developments. This technology quantitatively solves many problems in medicine, such as increased workload, delayed diagnosis, and treatment processes. It has been seen in the literature that this technology, which has a wide range of applications in medicine, also has a place in medical pathology. The main purpose of this study is to provide a histomorphological classification of colon polyps in the medical pathology department with high accuracy in a short time. Besides accelerating the diagnosis and treatment process, it is desired to facilitate the workload of the pathology department. This study is based on the recognition of colon preparation images that come to the pathology department by using the image recognition techniques of artificial intelligence. VGG19, DenseNet201, and EfcientNetB7 models, which are convolutional neural networks (CNN) models, were used. A model based on a concatenation ensemble of 3 CNN architectures was also used in this study. Within the scope of the study, a total of 515 preparation images, including normal epithelium, tubular adenoma and villous adenomas of the colon, were procured as data and introduced to artificial intelligence, and the diagnoses of these preparations were estimated histomorphologically by artificial intelligence. As a result of the study, an accuracy rate of 94.17% was achieved.

Comparative Analysis of Machine Learning Models for Image Detection of Colonic Polyps vs. Resected Polyps.

(1) Background: Colon polyps are common protrusions in the colon's lumen, with potential risks of developing colorectal cancer. Early detection and intervention of these polyps are vital for reducing colorectal cancer incidence and mortality rates. This research aims to evaluate and compare the performance of three machine learning image classification models' performance in detecting and classifying colon polyps. (2) Methods: The performance of three machine learning image classification models, Google Teachable Machine (GTM), Roboflow3 (RF3), and You Only Look Once version 8 (YOLOv8n), in the detection and classification of colon polyps was evaluated using the testing split for each model. The external validity of the test was analyzed using 90 images that were not used to test, train, or validate the model. The study used a dataset of colonoscopy images of normal colon, polyps, and resected polyps. The study assessed the models' ability to correctly classify the images into their respective classes using precision, recall, and F1 score generated from confusion matrix analysis and performance graphs. (3) Results: All three models successfully distinguished between normal colon, polyps, and resected polyps in colonoscopy images. GTM achieved the highest accuracies: 0.99, with consistent precision, recall, and F1 scores of 1.00 for the 'normal' class, 0.97-1.00 for 'polyps', and 0.97-1.00 for 'resected polyps'. While GTM exclusively classified images into these three categories, both YOLOv8n and RF3 were able to detect and specify the location of normal colonic tissue, polyps, and resected polyps, with YOLOv8n and RF3 achieving overall accuracies of 0.84 and 0.87, respectively. (4) Conclusions: Machine learning, particularly models like GTM, shows promising results in ensuring comprehensive detection of polyps during colonoscopies.

S731 Diagnostic Accuracy of Artificial Intelligence in Classification of Colonic Polyps – A Real World Prospective Study

S731 Diagnostic Accuracy of Artificial Intelligence in Classification of Colonic Polyps – A Real World Prospective Study

A New Method Based on Convolutional Neural Networks and Discrete Wavelet Transform for Detection, Classification and Tracking of Colon Polyps in Colonoscopy Videos

A New Method Based on Convolutional Neural Networks and Discrete Wavelet Transform for Detection, Classification and Tracking of Colon Polyps in Colonoscopy Videos

Automated Colonic Polyp Detection and Classification Enabled Northern Goshawk Optimization with Deep Learning

The major mortality factor relevant to the intestinal tract is the growth of tumorous cells (polyps) in various parts. More specifically, colonic polyps have a high rate and are recognized as a precursor of colon cancer growth. Endoscopy is the conventional technique for detecting colon polyps, and considerable research has proved that automated diagnosis of image regions that might have polyps within the colon might be used to help experts for decreasing the polyp miss rate. The automated diagnosis of polyps in a computer-aided diagnosis (CAD) method is implemented using statistical analysis. Nowadays, Deep Learning, particularly through Convolution Neural networks (CNN), is broadly employed to allow the extraction of representative features. This manuscript devises a new Northern Goshawk Optimization with Transfer Learning Model for Colonic Polyp Detection and Classification (NGOTL-CPDC) model. The NGOTL-CPDC technique aims to investigate endoscopic images for automated colonic polyp detection. To accomplish this, the NGOTL-CPDC technique comprises of adaptive bilateral filtering (ABF) technique as a noise removal process and image pre-processing step. Besides, the NGOTL-CPDC model applies the Faster SqueezeNet model for feature extraction purposes in which the hyperparameter tuning process is performed using the NGO optimizer. Finally, the fuzzy Hopfield neural network (FHNN) method can be employed for colonic poly detection and classification. A widespread simulation analysis is carried out to ensure the improved outcomes of the NGOTL-CPDC model. The comparison study demonstrates the enhancements of the NGOTL-CPDC model on the colonic polyp classification process on medical test images.

Performance and comparison of artificial intelligence and human experts in the detection and classification of colonic polyps

ObjectiveThe main aim of this study was to analyze the performance of different artificial intelligence (AI) models in endoscopic colonic polyp detection and classification and compare them with doctors with different experience.MethodsWe searched the studies on Colonoscopy, Colonic Polyps, Artificial Intelligence, Machine Learning, and Deep Learning published before May 2020 in PubMed, EMBASE, Cochrane, and the citation index of the conference proceedings. The quality of studies was assessed using the QUADAS-2 table of diagnostic test quality evaluation criteria. The random-effects model was calculated using Meta-DISC 1.4 and RevMan 5.3.ResultsA total of 16 studies were included for meta-analysis. Only one study (1/16) presented externally validated results. The area under the curve (AUC) of AI group, expert group and non-expert group for detection and classification of colonic polyps were 0.940, 0.918, and 0.871, respectively. AI group had slightly lower pooled specificity than the expert group (79% vs. 86%, P < 0.05), but the pooled sensitivity was higher than the expert group (88% vs. 80%, P < 0.05). While the non-experts had less pooled specificity in polyp recognition than the experts (81% vs. 86%, P < 0.05), and higher pooled sensitivity than the experts (85% vs. 80%, P < 0.05).ConclusionThe performance of AI in polyp detection and classification is similar to that of human experts, with high sensitivity and moderate specificity. Different tasks may have an impact on the performance of deep learning models and human experts, especially in terms of sensitivity and specificity.

Artificial Intelligence in Endoscopy.

Artificial intelligence (AI) is rapidly developing in various medical fields, and there is an increase in research performed in the field of gastrointestinal (GI) endoscopy. In particular, the advent of convolutional neural network, which is a class of deep learning method, has the potential to revolutionize the field of GI endoscopy, including esophagogastroduodenoscopy (EGD), capsule endoscopy (CE), and colonoscopy. A total of 149 original articles pertaining to AI (27 articles in esophagus, 30 articles in stomach, 29 articles in CE, and 63 articles in colon) were identified in this review. The main focuses of AI in EGD are cancer detection, identifying the depth of cancer invasion, prediction of pathological diagnosis, and prediction of Helicobacter pylori infection. In the field of CE, automated detection of bleeding sites, ulcers, tumors, and various small bowel diseases is being investigated. AI in colonoscopy has advanced with several patient-based prospective studies being conducted on the automated detection and classification of colon polyps. Furthermore, research on inflammatory bowel disease has also been recently reported. Most studies of AI in the field of GI endoscopy are still in the preclinical stages because of the retrospective design using still images. Video-based prospective studies are needed to advance the field. However, AI will continue to develop and be used in daily clinical practice in the near future. In this review, we have highlighted the published literature along with providing current status and insights into the future of AI in GI endoscopy.

Application of artificial intelligence for detection and classification of colon polyps

Colorectal cancer is one of the most common cancers worldwide, and colonoscopy has proven to be a preferable modality for screening and surveillance of colorectal cancer. This review discusses the clinical application of artificial intelligence (AI) and computer-aided diagnosis for automated colonoscopic detection and diagnosis of colorectal polyps for better understanding of the application of AI-based computer-aided diagnosis systems especially in terms of machine learning, deep learning and convolutional neural network for screening and surveillance of colorectal cancer.

Computational learning of features for automated colonic polyp classification

Shape, texture, and color are critical features for assessing the degree of dysplasia in colonic polyps. A comprehensive analysis of these features is presented in this paper. Shape features are extracted using generic Fourier descriptor. The nonsubsampled contourlet transform is used as texture and color feature descriptor, with different combinations of filters. Analysis of variance (ANOVA) is applied to measure statistical significance of the contribution of different descriptors between two colonic polyps: non-neoplastic and neoplastic. Final descriptors selected after ANOVA are optimized using the fuzzy entropy-based feature ranking algorithm. Finally, classification is performed using Least Square Support Vector Machine and Multi-layer Perceptron with five-fold cross-validation to avoid overfitting. Evaluation of our analytical approach using two datasets suggested that the feature descriptors could efficiently designate a colonic polyp, which subsequently can help the early detection of colorectal carcinoma. Based on the comparison with four deep learning models, we demonstrate that the proposed approach out-performs the existing feature-based methods of colonic polyp identification.

Colonoscopic Polyp Classification Using Local Shape and Texture Features

In this paper, a method is proposed for colonic polyp classification which can perform a virtual biopsy for assessing the stage of malignancy in polyps. Geometry, texture, and colour of a polyp give sufficient cue of its nature. The proposed framework characterizes geometry or shape of a polyp by pyramid histogram of oriented gradient (PHOG) features. To encapsulate the texture of the polyp surface, a fractal weighted local binary pattern (FWLBP) descriptor is employed, which is robust to affine transformation. It is also partially robust to illumination variations which is generally encountered during endoscopy. The optimal feature fusion is done using a feature ranking algorithm based on fuzzy entropy. Finally, to evaluate the classification performance of the proposed model, kernel-based support vector machines (SVM) and RUSBoosted tree are used. Experimental results carried on two databases clearly indicate that the proposed method can be used in the colonoscopic polyps classification. The proposed method can give polyp classification accuracies of 90.12% and 84.1%, and AUC of 0.91 and 0.92 for publicly available database and our own database, respectively.

An improved deep learning approach and its applications on colonic polyp images detection

BackgroundColonic polyps are more likely to be cancerous, especially those with large diameter, large number and atypical hyperplasia. If colonic polyps cannot be treated in early stage, they are likely to develop into colon cancer. Colonoscopy is easily limited by the operator’s experience, and factors such as inexperience and visual fatigue will directly affect the accuracy of diagnosis. Cooperating with Hunan children’s hospital, we proposed and improved a deep learning approach with global average pooling (GAP) in colonoscopy for assisted diagnosis. Our approach for assisted diagnosis in colonoscopy can prompt endoscopists to pay attention to polyps that may be ignored in real time, improve the detection rate, reduce missed diagnosis, and improve the efficiency of medical diagnosis.MethodsWe selected colonoscopy images from the gastrointestinal endoscopy room of Hunan children’s hospital to form the colonic polyp datasets. And we applied the image classification method based on Deep Learning to the classification of Colonic Polyps. The classic networks we used are VGGNets and ResNets. By using global average pooling, we proposed the improved approaches: VGGNets-GAP and ResNets-GAP.ResultsThe accuracies of all models in datasets exceed 98%. The TPR and TNR are above 96 and 98% respectively. In addition, VGGNets-GAP networks not only have high classification accuracies, but also have much fewer parameters than those of VGGNets.ConclusionsThe experimental results show that the proposed approach has good effect on the automatic detection of colonic polyps. The innovations of our method are in two aspects: (1) the detection accuracy of colonic polyps has been improved. (2) our approach reduces the memory consumption and makes the model lightweight. Compared with the original VGG networks, the parameters of our VGG19-GAP networks are greatly reduced.

Quest for the best endoscopic imaging modality for computer-assisted colonic polyp staging.

BACKGROUNDIt was shown in previous studies that high definition endoscopy, high magnification endoscopy and image enhancement technologies, such as chromoendoscopy and digital chromoendoscopy [narrow-band imaging (NBI), i-Scan] facilitate the detection and classification of colonic polyps during endoscopic sessions. However, there are no comprehensive studies so far that analyze which endoscopic imaging modalities facilitate the automated classification of colonic polyps. In this work, we investigate the impact of endoscopic imaging modalities on the results of computer-assisted diagnosis systems for colonic polyp staging.AIMTo assess which endoscopic imaging modalities are best suited for the computer-assisted staging of colonic polyps.METHODSIn our experiments, we apply twelve state-of-the-art feature extraction methods for the classification of colonic polyps to five endoscopic image databases of colonic lesions. For this purpose, we employ a specifically designed experimental setup to avoid biases in the outcomes caused by differing numbers of images per image database. The image databases were obtained using different imaging modalities. Two databases were obtained by high-definition endoscopy in combination with i-Scan technology (one with chromoendoscopy and one without chromoendoscopy). Three databases were obtained by high-magnification endoscopy (two databases using narrow band imaging and one using chromoendoscopy). The lesions are categorized into non-neoplastic and neoplastic according to the histological diagnosis.RESULTSGenerally, it is feature-dependent which imaging modalities achieve high results and which do not. For the high-definition image databases, we achieved overall classification rates of up to 79.2% with chromoendoscopy and 88.9% without chromoendoscopy. In the case of the database obtained by high-magnification chromoendoscopy, the classification rates were up to 81.4%. For the combination of high-magnification endoscopy with NBI, results of up to 97.4% for one database and up to 84% for the other were achieved. Non-neoplastic lesions were classified more accurately in general than non-neoplastic lesions. It was shown that the image recording conditions highly affect the performance of automated diagnosis systems and partly contribute to a stronger effect on the staging results than the used imaging modality.CONCLUSIONChromoendoscopy has a negative impact on the results of the methods. NBI is better suited than chromoendoscopy. High-definition and high-magnification endoscopy are equally suited.

Changing pathological diagnosis from hyperplastic polyp to sessile serrated adenoma: systematic review and meta-analysis.

The WHO published a new classification of colonic polyps in 2010, including the group of serrated polyps, which can be divided into hyperplastic polyps (HP), traditional serrated adenomas, and sessile serrated adenomas (SSA) or polyps. To assess the rate of re-diagnosis of HP to SSA and to look for possible predictors for changing the diagnosis. English Medical literature searches were performed for 'reassessment' OR 'reclassification' AND 'hyperplastic polyp' OR 'sessile serrated adenoma' till 31 January 2017. PRISMA guidelines for systematic reviews were followed. Studies that included a precise re-diagnosis of HP into SSA were included. We also looked for predictors of SSA diagnosis such as polyp location and size, patient sex and age, and synchronous advanced adenoma. Altogether, we found 220 eligible studies; 212 were excluded as they did not fulfill the inclusion criteria and we were left with eight studies including 2625 patients. The odds ratio for the number of polyps with changed pathological diagnosis from HP to SSA was 0.112 with 95% confidence interval (CI): 0.099-0.126 (P<0.0001) or 11.2%. Heterogeneity between studies was significant with Q=199.4, d.f. (Q)=9, P<0.0001, and I=95.486%. The odds ratio for changing the pathological diagnosis from HP to SSA for polyp proximal location and polyp size more than 5 mm were 4.401, 95% CI: 2.784-6.958, P<0.0001, and 8.336, 95% CI: 4.963-15.571, P<0.0001, respectively. Endoscopists and pathologists should be aware of the SSA diagnosis when finding HPs larger than 5 mm in the right colon. The diagnosis of HP in these cases should be reassessed by experienced gastrointestinal pathologists.

Functional Data Analysis of Spectroscopic Data with Application to Classification of Colon Polyps

In this study, two functional logistic regression models with functional principal component basis (FPCA) and functional partial least squares basis (FPLS) have been developed to distinguish precancerous adenomatous polyps from hyperplastic polyps for the purpose of classification and interpretation. The classification performances of the two functional models have been compared with two widely used multivariate methods, principal component discriminant analysis (PCDA) and partial least squares discriminant analysis (PLSDA). The results indicated that classification abilities of FPCA and FPLS models outperformed those of the PCDA and PLSDA models by using a small number of functional basis components. With substantial reduction in model complexity and improvement of classification accuracy, it is particularly helpful for interpretation of the complex spectral features related to precancerous colon polyps.

Diagnostic value of high-resolution micro-endoscopy for the classification of colon polyps.

To study a new imaging equipment, high-resolution micro-endoscopy (HRME), in the diagnosis and pathological classification of colon polyps. We selected 114 specimens of colon polyps, 30 of which were colon polyps with known pathological types and 84 that were prospective polyp specimens; 10 normal colon mucosa specimens served as controls. We obtained images of 30 colon polyp specimens with known pathological types using HRME and analyzed the characteristics of these images to develop HRME diagnostic criteria for different pathological types of colon polyps. Based on these criteria, we performed a prospective study of 84 colon polyp specimens using HRME and compared the results with those of the pathological examination to evaluate the diagnostic value of HRME in the pathological classification of different types of colon polyps. In the 30 cases of known pathological type of colon polyp samples, there were 21 cases of adenomatous polyps, which comprised nine cases of tubular adenoma, seven cases of villous adenoma and five cases of mixed adenomas. The nine cases of non-adenomatous polyps included four cases of inflammatory polyps and five cases of hyperplastic polyps five. Ten cases of normal colonic mucosa were confirmed pathologically. In a prospective study of 84 cases using HRME, 23 cases were diagnosed as inflammatory polyps, 11 cases as hyperplastic polyps, 18 cases as tubular adenoma, eight cases as villous adenoma and 24 cases as mixed adenomas. After pathological examination, 24 cases were diagnosed as inflammatory polyps, 11 cases as hyperplastic polyps, 19 cases as tubular adenoma, eight cases as villous adenoma and 22 cases as mixed adenomas. Compared with the pathological examinations, the sensitivities, specificities, accuracies, and positive and negative predictive values of HRME in diagnosing inflammatory polyps (87.5%, 96.7%, 94.0%, 91.3% and 95.1%), hyperplastic polyps (72.7%, 95.9%, 92.9%, 72.7% and 95.9%), tubular adenomas (73.7%, 93.8%, 89.3%, 77.8% and 92.4%), villous adenomas (75.0%, 97.4%, 95.2%, 75.0% and 97.4%), and mixed adenomas (75.0%, 93.3%, 88.1%, 81.8% and 90.3%) were relatively high. HRME has a relatively high diagnostic value in the pathological classification of colon polyps. Thus, it may be an alternative to confocal microendoscopy in lower-resource or community-based settings.

Exploring Deep Learning and Transfer Learning for Colonic Polyp Classification.

Recently, Deep Learning, especially through Convolutional Neural Networks (CNNs) has been widely used to enable the extraction of highly representative features. This is done among the network layers by filtering, selecting, and using these features in the last fully connected layers for pattern classification. However, CNN training for automated endoscopic image classification still provides a challenge due to the lack of large and publicly available annotated databases. In this work we explore Deep Learning for the automated classification of colonic polyps using different configurations for training CNNs from scratch (or full training) and distinct architectures of pretrained CNNs tested on 8-HD-endoscopic image databases acquired using different modalities. We compare our results with some commonly used features for colonic polyp classification and the good results suggest that features learned by CNNs trained from scratch and the “off-the-shelf” CNNs features can be highly relevant for automated classification of colonic polyps. Moreover, we also show that the combination of classical features and “off-the-shelf” CNNs features can be a good approach to further improve the results.

Local fractal dimension based approaches for colonic polyp classification.

This work introduces texture analysis methods that are based on computing the local fractal dimension (LFD; or also called the local density function) and applies them for colonic polyp classification. The methods are tested on 8 HD-endoscopic image databases, where each database is acquired using different imaging modalities (Pentax's i-Scan technology combined with or without staining the mucosa) and on a zoom-endoscopic image database using narrow band imaging. In this paper, we present three novel extensions to a LFD based approach. These extensions additionally extract shape and/or gradient information of the image to enhance the discriminativity of the original approach. To compare the results of the LFD based approaches with the results of other approaches, five state of the art approaches for colonic polyp classification are applied to the employed databases. Experiments show that LFD based approaches are well suited for colonic polyp classification, especially the three proposed extensions. The three proposed extensions are the best performing methods or at least among the best performing methods for each of the employed databases. The methods are additionally tested by means of a public texture image database, the UIUCtex database. With this database, the viewpoint invariance of the methods is assessed, an important features for the employed endoscopic image databases. Results imply that most of the LFD based methods are more viewpoint invariant than the other methods. However, the shape, size and orientation adapted LFD approaches (which are especially designed to enhance the viewpoint invariance) are in general not more viewpoint invariant than the other LFD based approaches.

Optimizing area under the ROC curve using semi-supervised learning

Receiver operating characteristic (ROC) analysis is a standard methodology to evaluate the performance of a binary classification system. The area under the ROC curve (AUC) is a performance metric that summarizes how well a classifier separates two classes. Traditional AUC optimization techniques are supervised learning methods that utilize only labeled data (i.e., the true class is known for all data) to train the classifiers. In this work, inspired by semi-supervised and transductive learning, we propose two new AUC optimization algorithms hereby referred to as semi-supervised learning receiver operating characteristic (SSLROC) algorithms, which utilize unlabeled test samples in classifier training to maximize AUC. Unlabeled samples are incorporated into the AUC optimization process, and their ranking relationships to labeled positive and negative training samples are considered as optimization constraints. The introduced test samples will cause the learned decision boundary in a multi-dimensional feature space to adapt not only to the distribution of labeled training data, but also to the distribution of unlabeled test data. We formulate the semi-supervised AUC optimization problem as a semi-definite programming problem based on the margin maximization theory. The proposed methods SSLROC1 (1-norm) and SSLROC2 (2-norm) were evaluated using 34 (determined by power analysis) randomly selected datasets from the University of California, Irvine machine learning repository. Wilcoxon signed rank tests showed that the proposed methods achieved significant improvement compared with state-of-the-art methods. The proposed methods were also applied to a CT colonography dataset for colonic polyp classification and showed promising results.11Matlab code of the proposed methods will be released on http://clinicalcenter.nih.gov/drd/summers.html once the paper is published.

Seeing Is Believing: Video Classification for Computed Tomographic Colonography Using Multiple-Instance Learning

In this paper, we present development and testing results for a novel colonic polyp classification method for use as part of a computed tomographic colonography (CTC) computer-aided detection (CAD) system. Inspired by the interpretative methodology of radiologists using 3-D fly-through mode in CTC reading, we have developed an algorithm which utilizes sequences of images (referred to here as videos) for classification of CAD marks. For each CAD mark, we created a video composed of a series of intraluminal, volume-rendered images visualizing the detection from multiple viewpoints. We then framed the video classification question as a multiple-instance learning (MIL) problem. Since a positive (negative) bag may contain negative (positive) instances, which in our case depends on the viewing angles and camera distance to the target, we developed a novel MIL paradigm to accommodate this class of problems. We solved the new MIL problem by maximizing a L2-norm soft margin using semidefinite programming, which can optimize relevant parameters automatically. We tested our method by analyzing a CTC data set obtained from 50 patients from three medical centers. Our proposed method showed significantly better performance compared with several traditional MIL methods.