Suspicious Regions Of Interest Research Articles

A Yolo-Based Model for Breast Cancer Detection in Mammograms

This work aims to implement an automated data-driven model for breast cancer detection in mammograms to support physicians’ decision process within a breast cancer screening or detection program. The public available CBIS-DDSM and the INbreast datasets were used as sources to implement the transfer learning technique on full-field digital mammography proprietary dataset. The proprietary dataset reflects a real heterogeneous case study, consisting of 190 masses, 46 asymmetries, and 71 distortions. Several Yolo architectures were compared, including YoloV3, YoloV5, and YoloV5-Transformer. In addition, Eigen-CAM was implemented for model introspection and outputs explanation by highlighting all the suspicious regions of interest within the mammogram. The small YoloV5 model resulted in the best developed solution obtaining an mAP of 0.621 on proprietary dataset. The saliency maps computed via Eigen-CAM have proven capable solution reporting all regions of interest also on incorrect prediction scenarios. In particular, Eigen-CAM produces a substantial reduction in the incidence of false negatives, although accompanied by an increase in false positives. Despite the presence of hard-to-recognize anomalies such as asymmetries and distortions on the proprietary dataset, the trained model showed encouraging detection capabilities. The combination of Yolo predictions and the generated saliency maps represent two complementary outputs for the reduction of false negatives. Nevertheless, it is imperative to regard these outputs as qualitative tools that invariably necessitate clinical radiologic evaluation. In this view, the model represents a trusted predictive system to support cognitive and decision-making, encouraging its integration into real clinical practice.

Evaluation of Multiparametric Magnetic Resonance Imaging and Correlation with Radical Prostatectomy Histopathology Specimen in Prostate Cancer.

Magnetic resonance imaging (MRI) has shown a great potential in the evaluation and management of prostate cancer. In this study, we would like to evaluate the benefit of multiparametric MRI in the detection and localization of prostate cancer by comparing it with the gold standard of histopathology from radical prostatectomy. In this single-centre prospective study, 90 consecutive patients underwent radical prostatectomy from November 2016 to May 2018. All patients first underwent multiparametric (mp)-MRI, and all suspicious regions of interest were delineated and recorded on a 5-point scale as defined in prostate imaging reporting and data system version 2 (PI-RADS V2) score. All radical prostatectomy specimens, acquired after robotic radical prostatectomy with extended pelvic lymphadenectomy, were sent for histopathological examination (HPE). The mean age of the 90 patients was 65.3years, and the mean serum prostate-specific antigen (PSA) was 16.9ng/ml. The sensitivity and specificity of mp-MRI in the detection of the corresponding region of interest (ROI) on HPE were 67.4% and 89.3% respectively. Positive predictive value (PPV), negative predictive value (NPV), and accuracy of mp-MRI in the detection of corresponding ROI on HPE were 86.3%, 73.3%, and 78.3% respectively. The mp-MRI detected 96.8% solitary lesions and 61.7% multifocal lesions on the corresponding ROI on HPE. Multiparametric MRI has an excellent specificity and reasonable sensitivity for the diagnosis of prostate cancer. It is a good modality for the detection of solitary tumours, higher-grade tumours, detection of seminal vesicle invasion and extracapsular extension and helps in the decision-making process before radical prostatectomy, focal therapy or selecting an appropriate candidate for active surveillance.

Abstract 458: Development and validation of convolutional neural network to identify regions of interest in lumpectomy margins using optical coherence tomography

Abstract Background/Objective: Optical coherence tomography (OCT) is the optical analog of high-frequency ultrasound and produces real-time, high-resolution images up to 2 mm deep. Multi-reader studies of OCT have shown differentiation of normal parenchyma from neoplasms, including DCIS and cancers, with >85% sensitivity and specificity. Intraoperative evaluation of breast lumpectomy margins (LMs) with OCT may help achieve negative margins at primary surgery and avoid re-excision. Artificial Intelligence can be trained to spot regions of interest (ROI) in OCT LM images suspicious for malignancy. The purpose of this study was to develop and validate an automated convolutional neural network (CNN) to screen OCT LM images for ROIs. Methods: Following IRB approval, LMs from 126 patients with ductal malignancy were OCT imaged. Images were compared to corresponding permanent histology and annotated by breast pathologists to create a training set of 25,000 control ROIs. A CNN algorithm was developed with 3 convolutional layers, a 3x3 kernel, and 3 fully connected layers to perform binary classification of images as “suspicious” or “non-suspicious” for malignancy. A weighted loss function was used to balance the training data for non-suspicious vs. suspicious images and to tune sensitivity and specificity. Once trained and weighted, the CNN was tested in a prospective study using OCT images of 29 LMs from 29 patients with biopsy-proven ductal carcinoma in situ (DCIS), invasive ductal carcinoma (IDC), or both. CNN results were compared to permanent histology. Results: The patient population was 61.5 ± 7.3 years old, 100% female, with Stage 0-1 disease. Disease included IDC (n=20), invasive lobular (n=2), DCIS (n=27), mixed (n=74), atypical ductal hyperplasia (n=24), as well as benign findings including atypical lobular hyperplasia (n=19), lymphatic invasion (n=13), lobular carcinoma in situ (n=12), usual ductal hyperplasia (n=35), and duct ectasia (n=17). Following primary surgery, LMs were scanned using OCT and images were CNN analyzed. Approximately 1.9 M OCT ROIs were assessed, identifying 101,099 suspicious ROIs. Three hundred and eighty-four (384) ROIs were correctly identified, yielding a 70% true positive and 5.2% false positive rate with 70% sensitivity and 96% specificity. The receiver operating curve is shown below. Conclusions: Automated analysis of OCT images using a trained CNN to identify ROIs suspicious for DCIS or IDC in LMs is feasible, demonstrating high concordance with permanent pathology. These findings indicate the utility of AI for screening OCT images with potential utilization for intraoperative evaluation of LMs. A pivotal prospective clinical trial will be necessary to evaluate breast specimens in real time to determine if this application may improve re-excision rates in lumpectomy. Citation Format: David Rempel, Andrew Berkeley, Chandandeep Nagi, Vladimir Pekar, Margaret Burns, Beryl Augustine, Alia Nazarullah, Ismail Jatoi, Kelly K. Hunt, Alastair Thompson, Savitri Krishnamurthy. Development and validation of convolutional neural network to identify regions of interest in lumpectomy margins using optical coherence tomography [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 458.

Quantitative analysis of in-TIPS thrombosis in abdominal CT

PurposeTo identify transjugular intrahepatic portosystemic shunt (TIPS) thrombosis in abdominal CT scans applying quantitative image analysis. Materials and methodsWe retrospectively screened 184 patients to include 20 patients (male, 8; female, 12; mean age, 60.7 ± 8.87 years) with (case, n = 10) and without (control, n = 10) in-TIPS thrombosis who underwent clinically indicated contrast-enhanced and unenhanced abdominal CT followed by conventional TIPS-angiography between 08/2014 and 06/2020. First, images were scored visually. Second, region of interest (ROI) based quantitative measurements of CT attenuation were performed in the inferior vena cava (IVC), portal vein and in four TIPS locations. Minimum, maximum and average Hounsfield unit (HU) values were used as absolute and relative quantitative features. We analyzed the features with univariate testing. ResultsSubjective scores identified in-TIPS thrombosis in contrast-enhanced scans with an accuracy of 0.667 – 0.833. Patients with in-TIPS thrombosis had significantly lower average (p < 0.001), minimum (p < 0.001) and maximum HU (p = 0.043) in contrast-enhanced images. The in-TIPS / IVC ratio in contrast-enhanced images was significantly lower in patients with in-TIPS thrombosis (p < 0.001). No significant differences were found for unenhanced images. Analyzing the visually most suspicious ROI with consecutive calculation of its ratio to the IVC, all patients with a ratio < 1 suffered from in-TIPS thrombosis (p < 0.001, sensitivity and specificity = 100%). ConclusionQuantitative analysis of abdominal CT scans facilitates the stratification of in-TIPS thrombosis. In contrast-enhanced scans, an in-TIPS / IVC ratio < 1 could non-invasively stratify all patients with in-TIPS thrombosis.

Improving Breast Cancer Classification using (SMOTE) Technique and Pectoral Muscle Removal in Mammographic Images

Computer-aided diagnosis methods are being developed to assist radiologists to improve the interpretation of mammograms for the detection and diagnose of breast cancer, reduce the errors and mistakes made by human beings. In addition, it provides a more accurate and reliable classification of benign and malignant abnormalities. In the mammogram diagnosis, the pectoral muscle appears in Mediolateral oblique views (MLO) of the right and left of the breast. Considering that, the pectoral muscle has the same density as the small suspicious masses in the image and can affect/bias the results of image processing methods. This paper presents a diagnosis method to detect an abnormality in mammograms automatically. Before abnormality identification, image-processing techniques are used to correctly segment the suspicious region-of-interest (ROI). The background of the mammograms has been darkened to distinguish the breast area from any blemishes or writings that will be removed. Then the breast area has been extracted after ignoring the empty regions around the breast in mammogram images. After that, the mammogram image is inverted and the inverted image is then subtracted from the initial image. For pectoral muscle removal, a region growing method using the K-means clustering method is used. Afterward, suspicious ROI is segmented utilizing the K-means with thresholding technique. To detect abnormalities in mammograms, shape-based features, moment invariants, and also fractal dimensions are extracted from the segmented ROI. The Mini-MIAS dataset is used to evaluate the proposed method and is predominately composed of benign samples with only a tiny percent of malignant samples. To accomplish far better classifier efficiency, the SMOTE algorithm is used to present new samples from the minority classes to get a balanced dataset. Random forest classifier utilized to classify the segmented region as benign and malignant. The experimental results obtained an accuracy of 97.1%, the sensitivity of 95.1%, and the achieved specificity is 98.5%.

Identifying prostate cancer in men with non-suspicious multi-parametric magnetic resonance imaging of the prostate.

To formulate clinical pathways for identifying clinically significant prostate cancer (csPC) and avoiding insignificant prostate cancer (isPC) in those without suspicious regions of interest on multi-parametric magnetic resonance imaging (mpMRI) of the prostate. A retrospective review identified patients with negative mpMRI who underwent subsequent transperineal prostate biopsy across two centres. Patient characteristics and association with biopsy results were evaluated using univariate and multivariate regression analyses. A total of 144 patients were identified as having negative mpMRI and undergoing subsequent transperineal prostate biopsy; 18% (25/144) of the cohort were found to have csPC. Logistic regression analysis failed to identify statistically significant predictive factors. In this cohort, if all patients with prostate-specific antigen > 3.0 were biopsied the least amount of csPC is missed, at 20% (5/25) however all isPC would be diagnosed. The least amount of isPC is diagnosed with a biopsy threshold of >15% from the European Randomized Study of Screening for Prostate Cancer calculator with 20% (5/25) of isPC diagnoses made however only 10.5% (2/19) csPC would be diagnosed. A biopsy threshold of >5% risk reduces the number of csPC missed to 37% (7/19) however increases isPC diagnoses to 54% (13/24) of the population. False-negative rates of prostate MRI for csPC are significant within our cohort at 18%. The decision to biopsy should be made in conjunction with a risk profile acceptable by the patient and clinician. The current study demonstrates that there is a need to balance the risk of missing csPC and harm of diagnosing isPC.

Semi-automated PIRADS scoring via mpMRI analysis.

Purpose: Prostate cancer (PCa) is the most common solid organ cancer and second leading cause of death in men. Multiparametric magnetic resonance imaging (mpMRI) enables detection of the most aggressive, clinically significant PCa (csPCa) tumors that require further treatment. A suspicious region of interest (ROI) detected on mpMRI is now assigned a Prostate Imaging-Reporting and Data System (PIRADS) score to standardize interpretation of mpMRI for PCa detection. However, there is significant inter-reader variability among radiologists in PIRADS score assignment and a minimal input semi-automated artificial intelligence (AI) system is proposed to harmonize PIRADS scores with mpMRI data. Approach: The proposed deep learning model (the seed point model) uses a simulated single-click seed point as input to annotate the lesion on mpMRI. This approach is in contrast to typical medical AI-based approaches that require annotation of the complete lesion. The mpMRI data from 617 patients used in this study were prospectively collected at a major tertiary U.S. medical center. The model was trained and validated to classify whether an mpMRI image had a lesion with a PIRADS score greater than or equal to PIRADS 4. Results: The model yielded an average receiver-operator characteristic (ROC) area under the curve (ROC-AUC) of 0.704 over a 10-fold cross-validation, which is significantly higher than the previously published benchmark. Conclusions: The proposed model could aid in PIRADS scoring of mpMRI, providing second reads to promote quality as well as offering expertise in environments that lack a radiologist with training in prostate mpMRI interpretation. The model could help identify tumors with a higher PIRADS for better clinical management and treatment of PCa patients at an early stage.

P.603 Aggression, hostility and stress in relation to quality of life in breast cancer patients

While magnetic resonance imaging-ultrasound fusion targeted biopsy allows for improved detection of clinically significant prostate cancer, a concerning amount of clinically significant disease is still missed. We hypothesized that a number of these misses are due to the learning curve associated with magnetic resonance imaging-ultrasound fusion targeted biopsy. We report the results of repeat magnetic resonance imaging-ultrasound fusion targeted biopsy in men with continued suspicion for cancer and the institutional learning curve in the detection of clinically significant prostate cancer with time.We analyzed the records of 1,813 prostate biopsies in a prospectively acquired cohort of men who presented for prostate biopsy in a 4-year period. All men were offered prebiopsy magnetic resonance imaging and were assigned a maximum PI-RADS™ (Prostate Imaging Reporting and Data System version 2) score. Biopsy outcomes in men with a suspicious region of interest were compared. The relationship between time and clinically significant prostate cancer detection was analyzed.The clinically significant prostate cancer detection rate increased 26% with time in men with a PI-RADS 4/5 region of interest. On repeat magnetic resonance imaging-ultrasound fusion targeted biopsy in men with continued suspicion for cancer 53% of those with a PI-RADS 4/5 region of interest demonstrated clinically significant discordance from the initial magnetic resonance imaging-ultrasound fusion targeted biopsy compared to only 23% with a PI-RADS 1/2 region of interest. Significantly less clinically significant prostate cancer was missed or under graded in the most recent biopsies compared to the earliest biopsies.The high upgrade rate on repeat magnetic resonance imaging-ultrasound fusion targeted biopsy and the increasing cancer detection rate with time show the significant learning curve associated with magnetic resonance imaging-ultrasound fusion targeted biopsy. Men with low risk or negative biopsies with a persistent, concerning region of interest should be promptly rebiopsied. Improved targeting accuracy with operator experience can help decrease the number of missed cases of clinically significant prostate cancer.

Multiparametric Ultrasound for Prostate Cancer Detection and Localization: Correlation of B-mode, Shear Wave Elastography and Contrast Enhanced Ultrasound with Radical Prostatectomy Specimens.

Similar to multiparametric magnetic resonance imaging, multiparametric ultrasound represents a promising approach to prostate cancer imaging. We determined the diagnostic performance of B-mode, shear wave elastography and contrast enhanced ultrasound with quantification software as well as the combination, multiparametric ultrasound, for clinically significant prostate cancer localization using radical prostatectomy histopathology as the reference standard. From May 2017 to July 2017, 50 men with biopsy proven prostate cancer underwent multiparametric ultrasound before radical prostatectomy at 1 center. Three readers independently evaluated 12 anatomical regions of interest for the likelihood of clinically significant prostate cancer on a 5-point Likert scale for all separate ultrasound modalities and multiparametric ultrasound. A logistic linear mixed model was used to estimate diagnostic performance for the localization of clinically significant prostate cancer (any tumor with Gleason score 3 + 4 = 7 or greater, tumor volume 0.5 ml or greater, extraprostatic extension or stage pN1) using a Likert score of 3 or greater and 4 or greater as the threshold. To detect the index lesion the readers selected the 2 most suspicious regions of interest. A total of 48 men were included in the final analysis. The region of interest specific sensitivity of multiparametric ultrasound (Likert 3 or greater) for clinically significant prostate cancer was 74% (95% CI 67-80) compared to 55% (95% CI 47-63), 55% (95% CI 47-63) and 59% (95% CI 51-67) for B-mode, shear wave elastography and contrast enhanced ultrasound, respectively. Multiparametric ultrasound sensitivity was significantly higher for Likert thresholds and all different clinically significant prostate cancer definitions (all p <0.05). Multiparametric ultrasound improved the detection of index lesion prostate cancer. Multiparametric ultrasound of the prostate, consisting of B-mode, shear wave elastography and contrast enhanced ultrasound with parametric maps, improved localization and index lesion detection of clinically significant prostate cancer compared to single ultrasound modalities, yielding good sensitivity.

Classification of Lung Nodules Using CT Images Based on Texture Features and Fractal Dimension Transformation

A new computer-aided detection (CAD) system for lung nodule detection and selection in computed tomography scans is substantiated and implemented. The method consists of the following stages: preprocessing based on threshold and morphological filtration, the formation of suspicious regions of interest using a priori information, the detection of lung nodules by applying the fractal dimension transformation, the computation of informative texture features for identified lung nodules, and their classification by applying the SVM and AdaBoost algorithms. A physical interpretation of the proposed CAD system is given, and its block diagram is constructed. The simulation results based on the proposed CAD method demonstrate advantages of the new approach in terms of standard criteria, such as sensitivity and the false-positive rate.

The Institutional Learning Curve of Magnetic Resonance Imaging-Ultrasound Fusion Targeted Prostate Biopsy: Temporal Improvements in Cancer Detection in 4 Years

While magnetic resonance imaging-ultrasound fusion targeted biopsy allows for improved detection of clinically significant prostate cancer, a concerning amount of clinically significant disease is still missed. We hypothesized that a number of these misses are due to the learning curve associated with magnetic resonance imaging-ultrasound fusion targeted biopsy. We report the results of repeat magnetic resonance imaging-ultrasound fusion targeted biopsy in men with continued suspicion for cancer and the institutional learning curve in the detection of clinically significant prostate cancer with time. We analyzed the records of 1,813 prostate biopsies in a prospectively acquired cohort of men who presented for prostate biopsy in a 4-year period. All men were offered prebiopsy magnetic resonance imaging and were assigned a maximum PI-RADS™ (Prostate Imaging Reporting and Data System version 2) score. Biopsy outcomes in men with a suspicious region of interest were compared. The relationship between time and clinically significant prostate cancer detection was analyzed. The clinically significant prostate cancer detection rate increased 26% with time in men with a PI-RADS 4/5 region of interest. On repeat magnetic resonance imaging-ultrasound fusion targeted biopsy in men with continued suspicion for cancer 53% of those with a PI-RADS 4/5 region of interest demonstrated clinically significant discordance from the initial magnetic resonance imaging-ultrasound fusion targeted biopsy compared to only 23% with a PI-RADS 1/2 region of interest. Significantly less clinically significant prostate cancer was missed or under graded in the most recent biopsies compared to the earliest biopsies. The high upgrade rate on repeat magnetic resonance imaging-ultrasound fusion targeted biopsy and the increasing cancer detection rate with time show the significant learning curve associated with magnetic resonance imaging-ultrasound fusion targeted biopsy. Men with low risk or negative biopsies with a persistent, concerning region of interest should be promptly rebiopsied. Improved targeting accuracy with operator experience can help decrease the number of missed cases of clinically significant prostate cancer.

A cost-sensitive Bayesian combiner for reducing false positives in mammographic mass detection.

Mammography is the most widely used modality for early breast cancer detection. This work proposes a new computer-aided mass detection approach, in which a denoising method called BM3D is first applied to mammograms. Afterwards, using an adaptive segmentation algorithm, images are segmented to suspicious regions of interest (ROIs) and then a classifier is used to understand the features of true positive (TP) and false positive (FP) patterns. In this way, from selected suspicious ROIs, fractal dimension, texture and intensity features are extracted. Subsequently, a discretization approach followed by correlation-based feature selection (CFS) is combined with a genetic algorithm to obtain the most representative features. To neutralize the classifier's bias in favor of the major class in imbalanced datasets, an oversampling algorithm is used. In the next step, a cost-sensitive ensemble classifier based on a trainable combiner is proposed in order to reduce the number of FP samples. Finally, the presented method is validated on miniMIAS and INBreast datasets. The free-response receiver operating characteristic (FROC) analysis results prove the efficiency of the proposed approach. A sensitivity of 88% and false positive per image (FPpI) of 0.78 for miniMIAS and also a sensitivity of 86% and FPpI of 0.75 for INBreast dataset were obtained.

Computer-aided detection of pulmonary nodules using dynamic self-adaptive template matching and a FLDA classifier

Improving the performance of computer-aided detection (CAD) system for pulmonary nodules is still an important issue for its future clinical applications. This study aims to develop a new CAD scheme for pulmonary nodule detection based on dynamic self-adaptive template matching and Fisher linear discriminant analysis (FLDA) classifier. We first segment and repair lung volume by using OTSU algorithm and three-dimensional (3D) region growing. Next, the suspicious regions of interest (ROIs) are extracted and filtered by applying 3D dot filtering and thresholding method. Then, pulmonary nodule candidates are roughly detected with 3D dynamic self-adaptive template matching. Finally, we optimally select 11 image features and apply FLDA classifier to reduce false positive detections. The performance of the new method is validated by comparing with other methods through experiments using two groups of public datasets from Lung Image Database Consortium (LIDC) and ANODE09. By a 10-fold cross-validation experiment, the new CAD scheme finally has achieved a sensitivity of 90.24% and a false-positive (FP) of 4.54 FP/scan on average for the former dataset, and a sensitivity of 84.1% with 5.59 FP/scan for the latter. By comparing with other previously reported CAD schemes tested on the same datasets, the study proves that this new scheme can yield higher and more robust results in detecting pulmonary nodules.

Optimizing safety and accuracy of prostate biopsy.

The objective of this article is to examine the safety of prostate biopsy and discuss the emerging role of MRI-ultrasound fusion technology in improving diagnostic accuracy. Men undergoing prostate biopsy frequently experience minor complications, including hematospermia, hematuria, and infection. Quinolone-resistant bacteria are a growing concern; thus, transperineal access or modification of antibiotic prophylaxis based on local antibiograms is now used to avoid infectious complications.Multiparametric MRI allows visualization of many prostate cancers, and by fusing MRI with real-time ultrasound, a biopsy needle can be directed by a urologist into suspicious regions of interest. Using this new method, detection of clinically significant prostate cancer has increased and the incidence of falsely negative biopsies has decreased. Prostate biopsy is generally a safe procedure, and with attention to local patterns of antibiotic resistance, infectious complications can be minimized. MRI-ultrasound fusion has significantly improved the accuracy of prostate biopsy, allowing tracking and targeting not previously possible.

Use of multiparametric MRI of prostate in active surveillance cohort of patients with localized prostate cancer in large urology group setting.

129 Background: The aim of this study was to determine the feasibility of using mpMRI to monitor/reclassify patients with low grade Prostate Cancer (PCa) undergoing Active Surveillance in a large urology practice. Methods: This is a retrospective study reviewing the records of 28 total patients following active surveillance (AS) protocol. All patients have been diagnosed with biopsy proven prostate cancer (PCa) and have undergone at least one mpMRI as a means of surveillance. The median total PSA of the group at time of diagnosis was 5.93 ng/ml and the mean age of the group was 64 years old. Between the 28 patients, 38 mpMRI tests were conducted, consisting of T2-weighted, diffusion weighted, and dynamic contrast enhanced MRI. Targeted ‘cognitive’ TRUS-MRI guided biopsy was used to locate progression of the PCa and help to determine whether a patient should stay on AS versus choosing more aggressive, definitive treatment. Confirmatory random biopsy was also used to determine overall sensitivity and specificity of the mpMRI conducted for suspicious regions of interest (ROI). Results: The sensitivity and specificity for the group was 65% and 92% respectively. The positive predictive value for the group was 94% while the negative predictive value was 55%. mpMRI was able to detect all clinically significant cancers except in one case. It did not detect some small size lesions (≤5.5mm) of Gleason score 6 that did not have a relevant clinical value. In five of the 28 cases (18%), patients fit the criteria for reclassification (Gleason ≥7, ≥ 3 new positive cores) out of the AS group. Four of these five patients exhibited clear progression of PCa on mpMRI and elected for more aggressive definitive treatment . The progression of cancer in the fifth patient was not seen on mpMRI but only after biopsy. This patient elected to stay in AS because of his advanced age. Conclusions: Despite the limitation of small sample size, mpMRI can be used as an effective diagnostic tool in the AS cohort of a community practice. It can facilitate early detection and progression of suspicious lesion(s) towards clinically significant PCa, thus triggering the start of more aggressive treatment.

Automatic detection of abnormal mammograms in mammographic images

This paper proposes a detection method for abnormal mammograms in mammographic datasets based on the novel abnormality detection classifier (ADC) by extracting a few of discriminative features, first-order statistical intensities and gradients. As tumorous masses are often indistinguishable from the surrounding parenchyma, automatic mass detection on highly complex breast tissues has been a challenge. However, most tumor detection methods require extraction of a large number of textural features for further multiple computations. The study first investigates image preprocessing techniques for obtaining more accurate breast segmentation prior to mass detection, including global equalization transformation, denoising, binarization, breast orientation determination and the pectoral muscle suppression. After performing gray level quantization on the breast images segmented, the presented feature difference matrices could be created by five features extracted from a suspicious region of interest (ROI); subsequently, principal component analysis (PCA) is applied to aid the determination of feature weights. The experimental results show that applying the algorithm of ADC accompanied with the feature weight adjustments to detect abnormal mammograms has yielded prominent sensitivities of 88% and 86% on the two respective datasets. Comparing other automated mass detection systems, this study proposes a new method for fully developing a high-performance, computer-aided decision (CAD) system that can automatically detect abnormal mammograms in screening programs, especially when an entire database is tested.

The Feasibility of Multiparametric Magnetic Resonance Imaging for Targeted Biopsy Using Novel Navigation Systems to Detect Early Stage Prostate Cancer: The Preliminary Experience

The feasibility and diagnostic performance of multiparametric magnetic resonance imaging (mp-MRI) has to be proven further. In this study, we evaluate the role of mp-MRI for targeted biopsy of early stage prostate cancer (PCa). A total 32 consecutive patients with transrectal ultrasonography (TRUS)-guided biopsy-proven PCa meeting low-risk criteria and pursuing active surveillance were selected to undergo mp-MRI 3 Tesla (3T) with endorectal coil. Patients were divided then into three groups based on the method used to target the mp-MRI designated region of interest (ROI): Group 1 underwent TRUS-guided prostate biopsy using an MRI-based coordinate plan (cognitive targeting). Group 2 underwent MRI-targeted TRUS-guided prostate biopsy using MyLabTMTwice, which superimposed the archived MRI images onto the real-time ultrasonography image allowing targeted biopsy of the ROI (fusion targeting). Group 3 included selected patients who had an elevation in prostate-specific antigen levels, or patients followed after radiation therapy (two patients) for suspicious unifocal MRI lesion recurrence. These patients underwent MRI-guided biopsy of the suspicious ROI using the navigation system DynaTRIM. The cancer detection rate in group 1 was 33.3% (3 of 10 patients), while in group 2, it was significantly higher at 46.2%. The sensitivity and specificity for group 1 was 45.5% and 33.3%, vs 61.9% and 20.8% in group 2, respectively. The positive predictive value in group 1 was 50.0% vs 53.8% in group 2 (P=0.04). In group 3, the cancer detection rate was much higher (80%) than in group 2, (P=0.005) although the majority of these patients (7 of 10) had a previously diagnosed prostate cancer on TRUS-guided 12-core biopsy. Our preliminary experience of mp-MRI suggests the detection of early stage prostate cancer with low-risk features yields potential candidates for active surveillance or focal targeted therapy. The MRI-TRUS fusion system increases diagnostic yield compared with cognitive MRI-directed TRUS-guided biopsy.

ANALYSIS OF MAMMOGRAM FOR DETECTION OF BREAST CANCER USING WAVELET STATISTICAL FEATURES

Early detection of breast cancer increases the survival rate and increases the treatment options. One of the most powerful techniques for early detection of breast cancer is based on digital mammogram. A system can be developed for assisting the analysis of digital mammograms using log-Gabor wavelet statistical features. The proposed system involves three major steps called Pre-processing, Processing, and Feature extraction. In pre-processing, the digital mammogram can be de-noised using efficient decision-based algorithm. In processing stage, the suspicious Region of Interest (ROI) can be cropped and convolved with log-Gabor filter for four different orientations. Then gray level co-occurrence matrix (GLCM)can be constructed for log-Gabor filter output at four different orientations and from that first order statistical features and second order statistical features can be extracted to analyze whether the mammogram as normal or benign or malignant. The proposed method can allow the radiologist to focus rapidly on the relevant parts of the mammogram and it can increase the effectiveness and efficiency of radiology clinics.

Mammographic Segmentation Using WaveCluster

Segmentation of clinically relevant regions from potentially noisy images represents a significant challenge in the field of mammography. We propose novel approaches based on the WaveCluster clustering algorithm for segmenting both the breast profile in the presence of significant acquisition noise and segmenting regions of interest (ROIs) within the breast. Using prior manual segmentations performed by domain experts as ground truth data, we apply our method to 150 film mammograms with significant acquisition noise from the University of South Florida’s Digital Database for Screening Mammography. We then apply a similar segmentation procedure to detect the position and extent of suspicious regions of interest. Our approach was able to segment the breast profile from all 150 images, leaving minor residual noise adjacent to the breast in three. Performance on ROI extraction was also excellent, with 81% sensitivity and 0.96 false positives per image when measured against manually segmented ground truth ROIs. When not utilizing image morphology, our approach ran in linear time with the input size. These results highlight the potential of WaveCluster as a useful addition to the mammographic segmentation repertoire.

Computer aided analysis of breast MRI enhancement kinetics using mean shift c lustering and multifeature iterative region of interest selection

To evaluate automatic characterization of a breast MR lesion by its spatially coherent region of interest (ROI). The method delineated 247 enhancing lesions using Otsu thresholding after manually placing a sphere. Mean Shift Clustering subdivided each volume, based on features including pharmacokinetic parameters. An iteratively trained classifier to predict the most suspicious ROI (IsR) was used, to predict the malignancy likelihood of each lesion. Performance was evaluated using receiver operator characteristic (ROC) analysis, and compared with a previous prototype. IsR was compared with noniterative training. The effect of adding BI-RADS™ morphology (from a radiologist) to the classifier was investigated. The area under the ROC curve (AUC) was 0.83 (95% confidence interval [CI] of 0.77-0.88), and was 0.75 (95%CI = 0.68-0.81; P = 0.029) without pharmacokinetic features. IsR performed better than conventional selection, based on one feature (AUC 0.75, 95%CI = 0.68-0.81; P = 0.035). With morphology, the AUC was 0.84 (95%CI = 0.78-0.88) versus 0.82 without (P = 0.40). Breast lesions can be characterized by their most suspicious, contiguous ROI using multi-feature clustering and iterative training. Characterization was improved by including pharmacokinetic modeling, while in our experiments, including morphology did not improve characterization.