CT Colonography Images Research Articles

Morphometric analysis and tortuosity typing of the large intestine segments on computed tomography colonography with artificial intelligence

Background:Morphological properties such as length and tortuosity of the large intestine segments play important roles, especially in interventional procedures like colonoscopy. Objective:Using computed tomography (CT) colonoscopy images, this study aimed to examine the morphological features of the colon's anatomical sections and investigate the relationship of these sections with each other or with age groups. The shapes of the transverse colon were analyzed using artificial intelligence. Materials and Methods:The study was conducted as a two- and three-dimensional examination of CT colonography images of people between 40 and 80 years old, which were obtained retrospectively. An artificial intelligence algorithm (YOLOv8) was used for shape detection on 3D colon images. Results:160 people with a mean age of 89 men and 71 women included in the study was 57.79±8.55 and 56.55±6.60, respectively, and there was no statistically significant difference (p=0.24). The total colon length was 166.11±25.07 cm for men and 158.73±21.92 cm for women, with no significant difference between groups (p=0.12). As a result of the training of the model Precision, Recall, and mAP were found to be 0.8578, 0.7940, and 0.9142, respectively. Conclusion:The study highlights the importance of understanding the type and morphology of the large intestine for accurate interpretation of CT colonography results and effective clinical management of patients with suspected large intestine abnormalities. Furthermore, this study showed that 88.57% of the images in the test data set were detected correctly and that AI can play an important role in colon typing.

Drainage pattern of the splenic flexure vein and its accompanying arteries using three-dimensional computed tomography angiography: a single-centre study of 600 patients.

The splenic flexure has variable vascular anatomy, and the details of the venous forms are not known. In this study, we report the flow pattern of the splenic flexure vein (SFV) and the positional relationship between the SFV and arteries such as the accessory middle colic artery (AMCA). This was a single-centre study using preoperative enhanced CT colonography images of 600 colorectal surgery patients. CT images were reconstructed into 3D angiography. SFV was defined as a vein flowing centrally from the marginal vein of the splenic flexure visible on CT. AMCA was defined as the artery feeding the left side of the transverse colon, separate from the left branch of the middle colic artery. The SFV returned to the inferior mesenteric vein (IMV) in 494 cases (82.3%), the superior mesenteric vein in 51 cases (8.5%) and the splenic vein in seven cases (1.2%). The AMCA was present in 244 cases (40.7%). The AMCA branched from the superior mesenteric artery or its branches in 227 cases (93.0% of cases with existing AMCA). In the 552 cases in which the SFV returned to the IMV, superior mesenteric vein or splenic vein, the left colic artery was the most frequent artery accompanying the SFV (42.2%), followed by the AMCA (38.1%) and the left branch of the middle colic artery (14.3%). The most common flow pattern of the vein in the splenic flexure is from the SFV to IMV. The SFV is frequently accompanied by the left colic artery or AMCA.

Determination of lower radiation dose limit for automatic measurement of adipose tissue.

The purpose of this study was to determine the lower limit of radiation dose required to measure visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) volumes when a fat quantification and noise reduction techniques (NRTs) are combined. For this purpose, we utilized CT colonography (CTC) images taken at low doses and manually segmented VAT and SAT fat volumes as ground truth. In order to derive the acceptable precision of the measurements needed to estimate the lower limit of radiation dose, we estimated the effect of different positioning during CT scanning on fat measurements using manually segmented VAT and SAT against normal dose. As a result, the acceptable accuracy of SAT and VAT was found to be 94.5% and 85.2%, respectively. Using these thresholds, the lower radiation dose limit required to accurately measure SAT using 5.25-mm slice-thick images was 1.5mGy of size-specific dose estimates (SSDE), while the lower radiation dose limit required to accurately measure VAT was 0.4mGy of SSDE. The lower dose limit for SAT and VAT combined was 1.5mGy, which was equivalent to an estimated effective dose of 0.38mSv. Alternatively, without noise reduction, SAT could not achieve acceptable accuracy even for images with a slice thickness of 5.25mm, while VAT required noise reduction for images with a slice thickness of 1.25mm, but could achieve acceptable accuracy without noise reduction for images with a slice thickness of 5.25mm.

Evaluation of BMI-based tube voltage selection in CT colonography: A prospective comparison of low kV versus routine 120kV protocol.

To explore the value of individualized kVp selection based on the patient's body mass index (BMI, kg/m2 ) in CT colonography (CTC). Seventy-eight patients underwent two CTC scans: conventional 120kVp in supine position (Group A) with 30% Adaptive statistical iteration algorithm (ASIR-V) and BMI-based lower kV p in prone position (Group B): tube voltage was suggested by an experienced investigator according to the patient's body mass index (BMI; calculated as weight divided by height squared; kg/m (2)).70kV for BMI<23kg/m2 (Group B1, n=27), 80kV for 23≤BMI≤25kg/m2 (Group B2, n=21) and 100kV for BMI>25kg/m2 (Group B3, n=30). Group A, corresponding to the BMI value in Group B, was divided into A1, A2, and A3 subgroups for analysis. Groups B used ASIR-V of different weights (30%-90% ASIR-V). The Hounsfield Unit (HU) and SD values of the muscles and the intestinal cavity air were measured, and the signal-to-noise ratio (SNR) and the contrast-to-noise ratio (CNR) of images were calculated. Imaging quality was evaluated by two reviewers and statistically compared. The 120kV scans were preferred more than 50% of the time. All images had excellent quality with good consistency between reviewers (Kappa>0.75, p<0.05). The radiation dose was reduced in groups B1, B2 and B3 by 63.62%, 44.63%, and 32.14%, respectively, compared with group A (p<0.05). The SNR and CNR values between group A1/A2/A3 and B1/B2/B3+60%ASIR-V were not statistically significant (p<0.05). There was no statistically significant difference between the subjective scores of group B combined with 60%ASIR-V and group A (p>0.05). BMI-based individualized kV CTC imaging significantly reduces overall radiation dose while providing an equal image quality with the conventional 120kV.

Physical Simulator for Colonoscopy: A Modular Design Approach and Validation

Simulators for gastrointestinal (GI) endoscopy offer the opportunity to train and assess clinician skills in a low-risk environment. Physical simulators can enable direct instrument-to-organ interactions not provided by virtual platforms. However, they present scarce visual realism and limited variability in their anatomical conditions. Herein, we present an innovative and low-cost methodology for the design and fabrication of modular silicone colon simulators. The fabrication pipeline envisages parametric customization and development of 3D-printed molds for silicon pouring to obtain colon segments. The size of each colon segment is based on clinical data extracted from CT colonography images. Straight and curved segments are connected through silicone conjuncts to realize a customized and modular monolithic physical simulator. A 130 cm-long colon simulator prototype with assorted magnetically-connected polyps was fabricated and laid on a custom-made sensorized abdominal phantom. Content, face and construct validity of the designed simulator were assessed by 17 GI endoscopists. In summary, this work showed promising results for improving accessibility and flexibility of current colonoscopy physical simulators, paving the way for modular and personalized training programs.

Deep learning in CT colonography: differentiating premalignant from benign colorectal polyps

ObjectivesTo investigate the differentiation of premalignant from benign colorectal polyps detected by CT colonography using deep learning.MethodsIn this retrospective analysis of an average risk colorectal cancer screening sample, polyps of all size categories and morphologies were manually segmented on supine and prone CT colonography images and classified as premalignant (adenoma) or benign (hyperplastic polyp or regular mucosa) according to histopathology. Two deep learning models SEG and noSEG were trained on 3D CT colonography image subvolumes to predict polyp class, and model SEG was additionally trained with polyp segmentation masks. Diagnostic performance was validated in an independent external multicentre test sample. Predictions were analysed with the visualisation technique Grad-CAM++.ResultsThe training set consisted of 107 colorectal polyps in 63 patients (mean age: 63 ± 8 years, 40 men) comprising 169 polyp segmentations. The external test set included 77 polyps in 59 patients comprising 118 polyp segmentations. Model SEG achieved a ROC-AUC of 0.83 and 80% sensitivity at 69% specificity for differentiating premalignant from benign polyps. Model noSEG yielded a ROC-AUC of 0.75, 80% sensitivity at 44% specificity, and an average Grad-CAM++ heatmap score of ≥ 0.25 in 90% of polyp tissue.ConclusionsIn this proof-of-concept study, deep learning enabled the differentiation of premalignant from benign colorectal polyps detected with CT colonography and the visualisation of image regions important for predictions. The approach did not require polyp segmentation and thus has the potential to facilitate the identification of high-risk polyps as an automated second reader.Key Points• Non-invasive deep learning image analysis may differentiate premalignant from benign colorectal polyps found in CT colonography scans.• Deep learning autonomously learned to focus on polyp tissue for predictions without the need for prior polyp segmentation by experts.• Deep learning potentially improves the diagnostic accuracy of CT colonography in colorectal cancer screening by allowing for a more precise selection of patients who would benefit from endoscopic polypectomy, especially for patients with polyps of 6–9 mm size.

Small bowel to closest human body surface distance calculation through a custom-made software using CT-based datasets.

Screening of the gastrointestinal tract is imperative for the detection and treatment of physiological and pathological disorders in humans. Ingestible devices (e.g., magnetic capsule endoscopes) represent an alternative to conventional flexible endoscopy for reducing the invasiveness of the procedure and the related patient's discomforts. However, to properly design localization and navigation strategies for capsule endoscopes, the knowledge of anatomical features is paramount. Therefore, authors developed a semi-automatic software for measuring the distance between the small bowel and the closest human external body surface, using CT colonography images. In this study, volumetric datasets of 30 patients were processed by gastrointestinal endoscopists with the dedicated custom-made software and results showed an average distance of 79.29 ± 23.85 mm.

Preoperative T staging of advanced colorectal cancer by computed tomography colonography.

Accurate preoperative T staging is important when determining the treatment strategy for advanced colorectal cancer. We have previously reported the usefulness of preoperative T staging based on the spatial relationship of tumors and "bordering vessels" by computed tomography colonography (CTC) with multiplanar reconstruction (MPR). The aims of this study were to evaluate the external validity of this method and to determine whether there is a difference in the accuracy of T staging between the mesenteric and antimesenteric sides. The study subjects were 110 patients with colorectal cancer who underwent preoperative CTC and surgical resection from June 2016 to March 2018. Preoperative T stage was determined by CTC based on the relationship between the tumor and the bordering vessels and compared with the pathological T stage. The influence of tumor location, namely, whether the tumor was on the antimesenteric or mesenteric side, on preoperative T staging was assessed in 78 patients with colorectal cancer. Sensitivity, specificity, accuracy, positive, and negative predictive values were respectively, 65%, 91%, 83%, 76%, and 85% for T2 (n = 34); 76%, 82%, 81%, 50%, and 94% for T3 (n = 23); and 77%, 93%, 87%, 86%, and 88% for T4a disease (n = 39). Overall right answer rate was 83.3% (15/18) for the mesenteric side and 65% (39/60) for the antimesenteric side (n = 0.14). Diagnostic criteria based on the bordering vessels seen on CTC images with MPR are useful for T staging of colorectal cancer. However, the accuracy differs between the antimesenteric and mesenteric sides.

Efficacy of CO₂ Infusion for Preoperative Computed Tomographic Angiography with Computed Tomographic Colonography.

BackgroundComputed tomographic colonography (CTC) is useful for patients for whom colonoscopy may be difficult to perform and is widely employed to examine the vasculature prior to colorectal cancer surgery. Computed tomographic angiography (CTA) was shown to be beneficial intraoperatively to manipulate blood vessels and prevent vascular injury. Three-dimensional (3D)-CTA combined with CTC (3D-CTA with CTC) is useful for preoperative evaluations of the anatomy of mesenteric vessels, colon, and lymph nodes. We observed that when the intestine was dilated with carbon dioxide (CO2), the arteriovenous delineation was often more pronounced than without CO2. To clarify the effects of gas injection with and without CO2 on hemodynamics and vascular passage, we compared the effect of contrast for blood vessels.Material/MethodsThirty patients with resectable colorectal cancer who underwent a preoperative CT examination at our institution from January to October 2019 were study participants. Of these, 15 underwent 3D-CTA and 15 had 3D-CTA with CTC. Three board-certified radiologists independently and blindly evaluated 18 blood vessels. CT values for each blood vessel were measured on each image.ResultsCT values for 3D-CTA with CTC were significantly higher with CO2 than without CO2. The quality of 3D-CTA with CTC images for visualization of blood vessels was also significantly greater than that of 3D-CTA, especially those of arterial and intramesenteric venous systems.ConclusionsBased on the higher image quality and CT values obtained by 3D-CTA with CTC for vessels, compared with by 3D-CTA imaging, 3D-CTA with CTC imaging might be useful in evaluating colorectal cancers.

3D visualization cloud based model to detect and classify the polyps according to their sizes for CT colonography

In the medical field, Medical diagnosis using Computed tomography (CT) has become increasingly popular due to their non-invasive approach and quick overall turnaround time. 3D visualization for CT Colonography involves the assessment and diagnosis of a patient to find the presence of cancerous polyps in the colon, by taking a Computed Tomography scan of the patient, and evaluating the reports. This technique reduces evaluation time by allowing the doctors themselves to analyze the CT scans without the need for a radiologist to generate an initial report. This technique also avoids an invasive procedure on the patient. This paper gives the insightsof developing computer aided system for the detection of Polyps in CT Colonography images using the principles of Image Processing and the Deep Learning, specifically an ensemble of Convolutional Neural Networks using GoogleNet architecture and 3D reconstruction of the same. The accuracy achieved by the proposed system for region classification, region 1 polyp detection and region 2 polyp detection are 98.75%, 93.75% and 94.03% respectively, and their F1 Scores are 0.88, 0.82 and 0.84 respectively.

Improving Image Correlation and Differentiation of 3D Endoluminal Lesions in the Air Spaces Using a Novel Target Gray Level Mapping Technique: A Preliminary Study of Its Application to Computed Tomographic Colonography and Comparison with Traditional Surface Rendering Method.

PurposeTo improve the three dimensional (3D) and two dimensional (2D) image correlation and differentiation of 3D endoluminal lesions in the traditional surface rendering (SR) computed tomographic endoscopy (CTE), a target gray level mapping (TGM) technique is developed and applied to computed tomographic colonography (CTC) in this study.MethodsA study of sixty-two various endoluminal lesions from thirty patients (13 males, 17 females; age range 31–90 years) was approved by our institutional review board and evaluated retrospectively. The endoluminal lesions were segmented using gray level threshold. The marching cubes algorithm was used to detect isosurfaces in the segmented volumetric data sets. TGM allows users to interactively apply grey level mapping (GM) to region of interest (ROI) in the 3D CTC. Radiologists conducted the clinical evaluation and the resulting data were analyzed. ResultsTGM and GM are significantly superior to SR in terms of surface texture, 3D shape, the confidence of 3D to 2D, 2D to 3D image correlation, and clinical classification of endoluminal lesions (P < 0.01). The specificity and diagnostic accuracy of GM and TGM methods are significantly better than those of SR (P < 0.01). Moreover, TGM performs better than GM (specificity: 75.0–85.7% vs. 53.6–64.3%; accuracy: 88.7–93.5% vs. 77.4–83.9%). TGM is a preferable display mode for further localization and differentiation of a lesion in CTC navigation. ConclusionsCompared with only the spatial shape information in traditional SR of CTC images, the 3D shapes and gray level information of endoluminal lesions can be provided by TGM simultaneously. 3D to 2D image correlations are also increased and facilitated at the same time. TGM is less affected by adjacent colon surfaces than GM. TGM serves as a better way to improve the image correlation and differentiation of endoluminal lesions.

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

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. 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. 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.

Dual-energy computed tomography colonography using dual-layer spectral detector computed tomography: Utility of virtual monochromatic imaging for electronic cleansing

ObjectiveTo assess the utility of virtual monochromatic imaging (VMI) using a dual-layer spectral detector CT for electronic cleansing in fecal-tagging CT colonography (CTC). MethodsThis study included 35 patients who underwent fecal-tagging CTC with a dual-layer detector spectral CT scanner. Conventional images at 120 kVp and VMI at 40, 50, and 60 keV were reconstructed. Quantitative image quality parameters, i.e., tagging density and image noise, were compared and the visual image quality was scored on a four-point scale. We recorded the number of the colon segments with appropriate tagging density (≥300 HU) for each patient and used these data to compare the reconstructions. In addition, electronic cleansing performance was semi-quantitatively assessed using a four-point scale. ResultsThe mean tagging density on VMI was significantly higher than that on conventional 120 kVp images. The number of colon segments with appropriate tagging density on VMI were significantly higher than that on conventional 120 kVp images. There was no significant difference among the reconstructed images with respect to image noise. Scores for subjective image quality and electronic cleansing performance on VMI were significantly higher than those on conventional 120 kVp images. ConclusionWith dual-layer spectral detector CT, VMI can yield significantly better fecal-tagged CTC image quality and improve electronic cleansing performance.

Risk Factors Associated With Incomplete Colonoscopy Based on the Analysis of Computed Tomographic Colonography Findings.

This study aimed to identify the computed tomographic colonography (CTC) findings of incomplete colonoscopy compared with those of complete colonoscopy. The clinical data and CTC imaging data from January 2004 to December 2012 were retrospectively obtained at 2 different institutions and reviewed by the central review system. A total of 71 patients who underwent both videocolonoscopy and CTC were included in this study. The CTC findings and clinical data were evaluated for the completeness of colonoscopy. In the CTC analysis, differences in total colon length, abdominal circumference, and sigmoid colon diameter were statistically significant between both groups (P < 0.05). Body mass index (BMI) and height were identified as significant clinical factors influencing the completeness of colonoscopy. In multiple logistic regression tests, only BMI and sigmoid colon diameter were independent factors (P < 0.05). High BMI larger diameter of sigmoid colon was associated with incomplete colonoscopy based on CTC.

JOURNAL CLUB: Extracolonic Findings at CT Colonography: Systematic Review and Meta-Analysis.

The purpose of this study was to perform a systematic review and meta-analysis of published studies on CT colonography (CTC) in which extracolonic findings were assessed. A systematic review of studies of screening CTC and of CTC to evaluate symptoms (1994-June 2017) was conducted to estimate the rate of extra-colonic findings and associated additional workup recommendations. The primary outcome was potentially important extracolonic findings, defined as CT Colonography Imaging Reporting and Data System (C-RADS) category E4 or as having high clinical importance (if C-RADS was not used). Secondary outcomes included likely unimportant extracolonic findings (i.e., C-RADS category E3 or similar). Random-effects and meta-regression analyses were used to generate pooled estimates and to explore risk factors for extracolonic findings related to various cohort characteristics. Primary data were acquired from 44 studies (49,676 patients). The pooled rate of potentially important extracolonic findings was 4.9% (95% CI, 3.7-6.4%) with a high degree of heterogeneity (I2 = 95%). This estimate progressively declined over time (9% per year since 2006) and was significantly related to the reporting system (lower for C-RADS than for low, moderate, high clinical significance reporting). C-RADS-specific meta-analysis (32,746 patients) showed rates of potentially significant extracolonic findings in 2.8% (95% CI, 1.9-3.5%) of the cohort without symptoms and 5.2% (95% CI, 3.5-7.6%) of the cohort with symptoms and in 5.7% (95% CI, 3.3-9.8%) of seniors (≥ 65 years) versus 2.3% (95% CI, 1.2-4.5%) of those younger than 65 years. The overall pooled rates of recommended workup were 8.2% (95% CI, 6.6-10.1%) for all extracolonic findings and 4.0% (95% CI, 2.7-5.9%) for potentially important extracolonic findings. With use of the more robust C-RADS classification, potentially important extracolonic findings at CTC occur in less than 3% of cohorts without symptoms. For all extracolonic findings, the rate of suggested or recommended additional workup is approximately 8% but decreases to 4% for potentially important extracolonic findings.

Post-imaging colorectal cancer or interval cancer rates after CT colonography: a systematic review and meta-analysis

CT colonography is highly sensitive for colorectal cancer, but interval or post-imaging colorectal cancer rates (diagnosis of cancer after initial negative CT colonography) are unknown, as are their underlying causes. We did a systematic review and meta-analysis of post-CT colonography and post-imaging colorectal cancer rates and causes to address this gap in understanding. We systematically searched MEDLINE, Embase, and the Cochrane Central Register of Controlled Trials. We included randomised, cohort, cross-sectional, or case-control studies published between Jan 1, 1994, and Feb 28, 2017, using CT colonography done according to international consensus standards with the aim of detecting cancer or polyps, and reporting post-imaging colorectal cancer rates or sufficient data to allow their calculation. We excluded studies in which all CT colonographies were done because of incomplete colonoscopy or if CT colonography was done with knowledge of colonoscopy findings. We contacted authors of component studies for additional data where necessary for retrospective CT colonography image review and causes for each post-imaging colorectal cancer. Two independent reviewers extracted data from the study reports. Our primary outcome was prevalence of post-imaging colorectal cancer 36 months after CT colonography. We used random-effects meta-analysis to estimate pooled post-imaging colorectal cancer rates, expressed using the total number of cancers and total number of CT colonographies as denominators, and per 1000 person-years. This study is registered with PROSPERO, number CRD42016042437. 2977 articles were screened and 12 studies were eligible for analysis. These studies reported data for 19 867 patients (aged 18-96 years; of 11 590 with sex data available, 6532 [56%] were female) between March, 2002, and May, 2015. At a mean of 34 months' follow-up (range 3-128·4 months), CT colonography detected 643 colorectal cancers. 29 post-imaging colorectal cancers were subsequently diagnosed. The pooled post-imaging colorectal cancer rate was 4·42 (95% CI 3·03-6·42) per 100 cancers detected, corresponding to 1·61 (1·11-2·33) post-imaging colorectal cancers per 1000 CT colonographies or 0·64 (0·44-0·92) post-imaging colorectal cancers per 1000 person-years. Heterogeneity was low (I2=0%). 17 (61%) of 28 post-imaging colorectal cancers were attributable to perceptual error and were visible in retrospect. CT colonography does not lead to an excess of post-test cancers relative to colonoscopy within 3-5 years, and the low 5-year post-imaging colorectal cancer rate confirms that the recommended screening interval of 5 years is safe. Since most post-imaging colorectal cancers arise from perceptual errors, radiologist training and quality assurance could help to reduce post-imaging colorectal cancer rates. St Mark's Hospital Foundation and the UK National Institute for Health Research via the UCL/UCLH Biomedical Research Centre.

Electronic cleansing in computed tomography colonography using AT layer identification with integration of gradient directional second derivative and material fraction model

BackgroundIn computed tomography colonography images, electronic cleansing (EC) is applied to remove opacified residual materials, called fecal-tagging materials (FTM), using positive-contrast tagging agents and laxative to facilitate polyp detection.MethodsThe proposed EC, ECprop, integrates the gradient directional second derivative into material fraction model to preserve submerged soft tissue (ST) under FTM. Three-material fraction model is used to remove FTM and artifacts at air-tagging (AT) layers and T-junctions where air, ST, and FTM material meet simultaneously. Moreover, the proposed AT layer identification is used to distinguish AT layers from air-tissue-tagging (ATT) layers in order to preserve ATT layers during cleansing. The clinical evaluation on 467 3-Dimensional band view images was conducted by the abdominal radiologist using four grading levels of cleansing quality with five causes of low quality EC. The amount of the remaining artifacts at T-junctions was approximated from the results of ECprop. The results from ECprop were compared with the results from syngo.via Client 3.0 Software, ECsyngo, and the fast three-material modeling, ECprev, using the preference of the radiologist. Two-tailed paired Wilcoxon signed rank test is used to indicate statistical significance.ResultsThe average grade on cleansing quality is 2.89 out of 4. The artifacts at T-junctions from 86.94% of the test images can be removed, whereas artifacts at T-junctions from only 13.06% of the test images cannot be removed. For 13.06% of the test images, the results from ECprop are more preferable to the results from ECsyngo (p<0.008). For all the test images, the results from ECprop are more preferable to the results from ECprev (p<0.001). Finally, the visual assessment shows that ECprop can preserve ATT layers, submerged polyps and folds while ECprev can preserve only submerged folds but fails to preserve ATT layers.ConclusionFrom our implementation, ECprop can improve the performance of the existing EC, such that it can preserve ST, especially ATT layers and remove the artifacts at T-junctions which have never been proposed by any other methods before.

Measurement of smaller colon polyp in CT colonography images using morphological image processing.

Automated measurement of the size and shape of colon polyps is one of the challenges in Computed tomography colonography (CTC). The objective of this retrospective study was to improve the sensitivity and specificity of smaller polyp measurement in CTC using image processing techniques. A domain knowledge-based method has been implemented with hybrid method of colon segmentation, morphological image processing operators for detecting the colonic structures, and the decision-making system for delineating the smaller polyp-based on a priori knowledge. The method was applied on 45 CTC dataset. The key finding was that the smaller polyps were accurately measured. In addition to 6-9mm range, polyps of even <5mm were also detected. The results were validated qualitatively and quantitatively using both 2D MPR and 3D view. Implementation was done on a high-performance computer with parallel processing. It takes [Formula: see text]min for measuring the smaller polyp in a dataset of 500 CTC images. With this method, [Formula: see text] and [Formula: see text] were achieved. The domain-based approach with morphological image processing has given good results. The smaller polyps were measured accurately which helps in making right clinical decisions. Qualitatively and quantitatively the results were acceptable when compared to the ground truth at [Formula: see text].

Potential pitfalls in the anorectal region during CT colonography: A discussion and pictorial overview of common pitfalls

Common potential anorectal pitfalls are presented with accompanying CT colonography images. Although most lesions are benign, care must be taken in not missing an underlying cancer.

Heterogeneous data analysis: Online learning for medical-image-based diagnosis

Heterogeneous Data Analysis (HDA) is proposed to address a learning problem of medical image databases of Computed Tomographic Colonography (CTC). The databases are generated from clinical CTC images using a Computer-aided Detection (CAD) system, the goal of which is to aid radiologists' interpretation of CTC images by providing highly accurate, machine-based detection of colonic polyps. We aim to achieve a high detection accuracy in CAD in a clinically realistic context, in which additional CTC cases of new patients are added regularly to an existing database. In this context, the CAD performance can be improved by exploiting the heterogeneity information that is brought into the database through the addition of diverse and disparate patient populations. In the HDA, several quantitative criteria of data compatibility are proposed for efficient management of these online images. After an initial supervised offline learning phase, the proposed online learning method decides whether the online data are heterogeneous or homogeneous. Our previously developed Principal Composite Kernel Feature Analysis (PC-KFA) is applied to the online data, managed with HDA, for iterative construction of a linear subspace of a high-dimensional feature space by maximizing the variance of the non-linearly transformed samples. The experimental results showed that significant improvements in the data compatibility were obtained when the online PC-KFA was used, based on an accuracy measure for long-term sequential online datasets. The computational time is reduced by more than 93% in online training compared with that of offline training.