Mediolateral Views Research Articles

The radiographic appearance of the canine femoral nutrient foramen for dogs undergoing total hip replacement is consistent, with some variations.

This study aimed to document the radiographic appearance of the femoral nutrient foramina and the variation of this in dogs undergoing total hip replacement (THR). Our hypothesis was that the radiographic appearance of the foramen would be consistent with the previously described anatomy, with some variations. 89 client-owned dogs. Preoperative radiographs were retrospectively analyzed for dogs undergoing THR at a single referral center. The signalment of all dogs was recorded. Radiographs were retrospectively examined to describe the number, direction, appearance, and foraminal index of the femoral nutrient foramen. Radiographs of 89 dogs and 102 femurs were examined. In 73 cases, a single foramen was seen; in 19 cases, no foramen was visible; and in 10 cases, 2 foramina were visible. The median foraminal index was 33.1% (range, 26% to 55.3%). On the mediolateral view, 72 were of proximocaudal-to-distocranial orientation, 19 were proximocranial to distocaudal, and 1 was atypical. On the craniocaudal or ventrodorsal views, the foramen was seen as a focal round radiolucency in 65 cases, was curved or atypical in 13 cases, and was not visible in 14. Radiolucent lines across the cortices that do not fit these criteria should raise suspicion of a femoral fissure, particularly within the context of THR.

An Audit of Diagnostic Mammography Examination among Women Presenting with Breast Symptoms

Background: Mammography is one of the imaging modalities that has been in existence for a long period of time in investigating the breast for pathologies. Its use in imaging the breast can be for the purpose of screening and diagnosis. Breast screening allows for early detection of breast cancer. Early detection of breast can save lives. Despite its important imaging modality for breast, mammography is not readily available in this part of the world. Our institution just acquired one recently. Objectives: Therefore, the aim of our study is to document the pattern of mammographic findings and to determine the prevalence of breast pathologies (benign and malignant) among symptomatic patients referred for mammography. Methodology: This descriptive cross-sectional study comprises 121 symptomatic women who presented for mammographic examination at the Radiology Department of the University of Abuja Teaching Hospital, Gwagwalada from January 2016 to December 2018. Two basic views (craniocaudal and mediolateral views) of the breast were obtained using an EXR-650 mammographic machine. Some of the patients had complimentary ultrasound scans. Results: The mean age of the study population was 47.23± 8.25 with an age range of 32–77 years. The majority of the patients were in the age group 40–49 years. Most women in the study were premenopausal and the majority had no family history of breast cancer. Breast pain was the most common symptom patients presented with. Heterogenous fibro glandular tissue accounted for the majority of mammographic breast density. Age correlated positively with a pattern of breast density among the study population. Abnormal mammographic findings were seen in 67 (55.4%) patients among which benign mass was the most prevalent. Among patients who presented with breast pain majority had normal mammographic findings while for those with breast lump, benign breast mass accounted for 35.7% representing the majority and malignant mass accounted for 19.1%. The relationship between mammographic outcome and indication was statistically significant P = 0.0001. Conclusion: Mammography plays a pivotal role in the evaluation of the breast in patients who presented with symptoms as it can differentiate benign and malignant lesions of the breast. In this study, the use of mammography reveals various pathologies in which benign breast mass accounted for the majority of breast pathologies detected.

Comparison and interobserver reliability of three different methods for measuring the amount of the tibial tuberosity advancement in the preoperative planning of TTAT.

The goal of preoperative planning techniques for advancement of the tibial tuberosity is to determine the amount of advancement required to achieve a postoperative patellar tendon angle of 90° and to select the optimal wedge size to achieve this target. Three radiographic methods for determining the advancement distance for the tibial tuberosity were evaluated for comparability and interobserver reliability. Among the methods developed, we decided to include the common tangent method, the tibial anatomy-based method, and the Bielecki method. For all techniques, radiographs were taken in mediolateral projection with the knee joint flexed at 135°. Three observers with different levels of experience independently evaluated and scored the degree of osteoarthritis for each stifle, as previously described, and performed measurements of the amount of advancement of the tibial tuberosity on 33 stifles using common tangent method, tibial anatomy-based method, and Bielecki method. According to the results, the overall score for osteoarthritis in the mediolateral view was influenced by the experience of the observers, which contradicts the results from a previous study. Regarding the measurement methods used to assess advancement, poor interobserver reliability was found for common tangent method and Bielecki method, while only a slightly moderate interobserver agreement was found for tibial anatomy-based method. These results are inconsistent with data collected by Bielecki and colleagues. Moreover, measurements from common tangent method and tibial anatomy-based method were overlapping, as confirmed previously. Conversely, Bielecki method showed no agreement with the other methods included in the present study, with a significantly higher mean rank, probably due to its correction formula. Based on the results of the present study, tibial anatomy-based method has better interobserver reliability and is easier to perform according to the observers.

A YOLO-based AI system for classifying calcifications on spot magnification mammograms

ObjectivesUse of an AI system based on deep learning to investigate whether the system can aid in distinguishing malignant from benign calcifications on spot magnification mammograms, thus potentially reducing unnecessary biopsies.MethodsIn this retrospective study, we included public and in-house datasets with annotations for the calcifications on both craniocaudal and mediolateral oblique vies, or both craniocaudal and mediolateral views of each case of mammograms. All the lesions had pathological results for correlation. Our system comprised an algorithm based on You Only Look Once (YOLO) named adaptive multiscale decision fusion module. The algorithm was pre-trained on a public dataset, Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM), then re-trained and tested on the in-house dataset of spot magnification mammograms. The performance of the system was investigated by receiver operating characteristic (ROC) analysis.ResultsWe included 1872 images from 753 calcification cases (414 benign and 339 malignant) from CBIS-DDSM. From the in-house dataset, 636 cases (432 benign and 204 malignant) with 1269 spot magnification mammograms were included, with all lesions being recommended for biopsy by radiologists. The area under the ROC curve for our system on the in-house testing dataset was 0.888 (95% CI 0.868–0.908), with a sensitivity of 88.4% (95% CI 86.9–8.99%), specificity of 80.8% (95% CI 77.6–84%), and an accuracy of 84.6% (95% CI 81.8–87.4%) at the optimal cutoff value. Using the system with two views of spot magnification mammograms, 80.8% benign biopsies could be avoided.ConclusionThe AI system showed good accuracy for classification of calcifications on spot magnification mammograms which were all categorized as suspicious by radiologists, thereby potentially reducing unnecessary biopsies.

Reconstruction of Three-Dimensional Tibiofemoral Kinematics Using Single-Plane Fluoroscopy and a Personalized Kinematic Model

Model-based 3D/2D image registration using single-plane fluoroscopy is a common setup to determine knee joint kinematics, owing to its markerless aspect. However, the approach was subjected to lower accuracies in the determination of out-of-plane motion components. Introducing additional kinematic constraints with an appropriate anatomical representation may help ameliorate the reduced accuracy of single-plane image registration. Therefore, this study aimed to develop and evaluate a multibody model-based tracking (MbMBT) scheme, embedding a personalized kinematic model of the tibiofemoral joint for the measurement of tibiofemoral kinematics. The kinematic model was consisted of three ligaments and an articular contact mechanism. The knee joint activities in six volunteers during isolated knee flexion, lunging, and sit-to-stand motions were recorded with a biplane X-ray imaging system. The tibiofemoral kinematics determined with the MbMBT and mediolateral view fluoroscopic images were compared against those determined using biplane fluoroscopic images. The MbMBT was demonstrated to yield tibiofemoral kinematics with precision values in the range from 0.1 mm to 1.1 mm for translations and from 0.2° to 1.3° for rotations. The constraints provided by the kinematic model were shown to effectively amend the nonphysiological tibiofemoral motion and not compromise the image registration accuracy with the proposed MbMBT scheme.

Radiographic and Biochemical Evaluations of Postnatal Development of Bones of the Forelimb in Nigerian Indigenous Dogs.

Bone is a hard and resilient tissue structure that provides mechanical support for the body of an organism. This study was aimed to evaluate the radiographical and biochemical changes during postnatal bone development of forelimbs in Nigerian indigenous puppies from 2-week-old to 24-week-old. Six (6) Nigerian indigenous puppies were chemically restrained using chlorpromazine at 2 mg/kg I.M. and positioned using ropes and sandbags for mediolateral and craniocaudal views of bones of the forelimbs. At 2-week-old, radiographic evaluation of bones of the forelimbs were undertaken and continued bi-weekly up to 24-week-old. Serum samples were analysed from 4-week-old to 24-week-old at four weeks’ interval for mineral (electrolytes) using colometric method for calcium and phosphorus and flame photometry method for sodium and potassium. At 2-week-old, there was an ill-defined radiopaque secondary ossification centre at the proximal epiphysis and a small ovoid radiopaque structure at the distal epiphysis of the humerus. There was not any sign of secondary ossification centre at proximal and distal epiphyses of both radius and ulna. There was the presence of a small radiopaque (ill-defined) accessory carpal. 1st to 5th metacarpals were present together with the proximal, middle and distal phalanges of all digits except the 1st digit that has proximal and distal phalanges. From 20-week-old, fully developed carpals and bony structures were observed through 24-week-old where the commencement of fusion of apophyses of olecranon (proximal epiphysis of ulna) at the centre and acrophyses of metacarpals and phalanges. The mean values (mean ± SEM) from serum biochemistry of calcium, phosphate, sodium and potassium in mmol/L from this study were within the normal limits throughout the study. Conclusively, the complementary information from calcium, phosphate, sodium and potassium in this study showed that these elements are within the normal limit that suggests normal bone development during the study period, which stressed the importance of the knowledge on serial changes of bone development which could be useful in clinical practice.

Radiographic Determination of the Canine Elbow Joint Angle in Collimated Views

Abstract The mediolateral flexed, extended, or neutral elbow radiographic views are commonly used in clinical practice. However, there is currently no standardized methodology to accurately measure the elbow joint angle in mediolateral images that include only the elbow joint and surrounding tissues. The main aim of this work is to compare elbow joint angles obtained from mediolateral radiographs that include the complete arm and forearm of the dog, with angles measured in radiographs including only the elbow. Ninety mediolateral views of elbow joints were obtained from 50 canine thoracic limbs, with 39 joints <90º, 30 ≥90 - ≤120º and 21 >120º. Radiographs were centered on the elbow joint and include the shoulder and carpal joints. For each complete forelimb radiographic image, the elbow angle was measured using the methodology described in previous studies. Then, the digital images were cut to obtain only the joint and surrounding tissues, establishing a new set of anatomical landmarks to measure the joint angles: the lateral humeral epicondyle was used as an angular point, with the linking points being the nutritional orifice of the radius at the antebrachial interosseous space and the intersection point of the lateral supracondylar crest with the cranial humeral endosteum. There was a good agreement observed between the two elbow angle measurement methodologies. The intraclass correlation coefficient was statistically significant, with the lower limits of the 95% confidence interval (CI) at >0.75, and with zero being included in the standard error of the mean 95% confidence interval in the Bland-Altman test. This elbow angle measurement methodology based on anatomic landmarks next to the elbow joint is accurate and may be used for clinical and research purposes.

Outcome of mammography examination in asymptomatic women.

Background:Early detection of breast cancer is important in reducing mortality, morbidity, and high socioeconomic burden associated with it. Mammography is currently the primary imaging modality used as a screening tool to detect early breast cancer in women experiencing no symptoms as they are most curable in the early stage with availability of breast conservative therapies.Objective:This study aimed at determining the mammographic breast density patterns and outcome in asymptomatic women who presented for mammographic examination in Abuja.Materials and Methods:This descriptive cross-sectional study comprises of 113 asymptomatic women who presented for mammographic examination at the Radiology Department of University of Abuja Teaching Hospital, Gwagwalada from March 2015 to December 2018. Two basic views (craniocaudal and mediolateral views) of the breast were obtained using EXR-650 mammographic machine.Results:The mean age of study population was 40.72 ± 10.45 years with age range of 35 and 65 years. Base on mammographic breast density, breast imaging, reporting, and data system 1 and 2 were the most prevalent. There was a positive correlation between mammographic breast density and age of respondents. This relationship was statistically significant (Pearson correlation = 0.56, P = 0.000). The mammographic outcome among asymptomatic women who had mammographic examination was negative in 69 (61.1%) women and positive in 44 (38.9%). The positive outcome noted in mammograms of women examined was: benign mass in 18 (15.9%) women; 9 (8.0%) had benign calcification; 7 (6.2%) showed architectural distortion; 5 (4.4%) was inconclusive; focal asymmetry in 3 (2.6%); and suspicious mass in 2 (1.8%).Conclusion:In this study, screening of women reveals various benign and malignant breast pathologies which necessitate early interventions.

Association of the Differences in Average Glandular Dose with Breast Cancer Risk

Objectives. To compare the differences in normalized average glandular dose (NAGD) between the breasts of healthy subjects and those of cancer patients and to determine if the NAGD difference is associated with breast cancer risk and improves breast cancer classification. Materials and Methods. Craniocaudal view and mediolateral view full-field digital mammography (FFDM) images were obtained from 1682 healthy subjects whose breasts were categorized as Breast Imaging-Reporting and Data System (BI-RADS) I or II and from 811 biopsy-confirmed unilateral breast cancer patients whose breasts on the contralateral side were category I or II. Both populations were randomized into training and test sets. Multivariate logistic regression analysis was used to build the breast cancer risk assessment model, and the area under the receiver operating characteristic curve ( A z ) was used to evaluate the model. Twenty-two breast cancer patients who were originally categorized as BI-RADS I or II for both breasts, but were diagnosed with unilateral biopsy-confirmed breast cancer subsequently, were included to validate the model. Results. The NAGD differences in both FFDM images between tumor-bearing breasts and the healthy breasts of patients were significantly higher than those in healthy subjects ( P < 0.001 ). The model with NAGD differences had a higher A z value than the model without NAGD differences. While there was no NAGD differences between originally healthy breasts of breast cancer patients, significant NAGD differences between now tumor-bearing breasts and the then previously healthy breasts were found in both FFDM images. Conclusions. NAGD differences between both breasts can be included in the breast cancer risk assessment model to evaluate breast cancer risk.

Automated Breast Cancer Detection in Digital Mammograms of Various Densities via Deep Learning.

Mammography plays an important role in screening breast cancer among females, and artificial intelligence has enabled the automated detection of diseases on medical images. This study aimed to develop a deep learning model detecting breast cancer in digital mammograms of various densities and to evaluate the model performance compared to previous studies. From 1501 subjects who underwent digital mammography between February 2007 and May 2015, craniocaudal and mediolateral view mammograms were included and concatenated for each breast, ultimately producing 3002 merged images. Two convolutional neural networks were trained to detect any malignant lesion on the merged images. The performances were tested using 301 merged images from 284 subjects and compared to a meta-analysis including 12 previous deep learning studies. The mean area under the receiver-operating characteristic curve (AUC) for detecting breast cancer in each merged mammogram was 0.952 ± 0.005 by DenseNet-169 and 0.954 ± 0.020 by EfficientNet-B5, respectively. The performance for malignancy detection decreased as breast density increased (density A, mean AUC = 0.984 vs. density D, mean AUC = 0.902 by DenseNet-169). When patients’ age was used as a covariate for malignancy detection, the performance showed little change (mean AUC, 0.953 ± 0.005). The mean sensitivity and specificity of the DenseNet-169 (87 and 88%, respectively) surpassed the mean values (81 and 82%, respectively) obtained in a meta-analysis. Deep learning would work efficiently in screening breast cancer in digital mammograms of various densities, which could be maximized in breasts with lower parenchyma density.

Additive value of exposure parameters for breast cancer diagnosis in digital mammography.

To investigate the effect of different breast lesions on exposure parameters in digital mammography and to determine whether the exposure parameters can additively improve diagnostic efficiency. Craniocaudal view and mediolateral view full-field digital mammography images from 982 women with unilateral lesions (341 with malignant lesions, 189 with benign lesions, and 452 healthy women) obtained at Nanfang Hospital were reviewed. Differences in exposure parameters (tube voltage and load, breast thickness (BT), and average glandular dose (AGD)) between breasts were calculated. The relationships between parameter differences and lesion size were explored. A logistic regression model was used based on the AGD and BT differences, and the area under the receiver operating characteristic curve (AUC) was used to assess the performance of these parameters in differentiating malignant from benign and healthy subjects. Independently, data from 129 women (82 with malignant and 47 with benign lesions) treated at Sun Yat-sen Memorial Hospital were collected to validate the model. Differences in tube voltage and load, BT, and AGD between breasts were significantly greater in the malignant subjects than benign (p < 0.05) and healthy subjects (p < 0.05). The AUCs for the comparisons of malignant vs. healthy subjects, malignant vs. benign subjects, and benign vs. healthy subjects were 0.77 ± 0.02, 0.72 ± 0.02, and 0.57 ± 0.02, respectively. The model combining the exposure parameters with the BI-RADS category resulted in a higher AUC (0.910 ± 0.03) compared with physician diagnosis alone (0.820 ± 0.04) for differentiating between malignant and benign lesions. Exposure parameters additively improved diagnostic accuracy for breast cancer and yielded more reliable results. • Differences in kVp, mAs, BT, and AGD between breasts were significantly greater in the malignant subjects than benign and healthy subjects. • The model combining exposure parameters with the BI-RADS category resulted in a higher AUC compared with the physician's diagnosis for differentiating between malignant and benign lesions. • Exposure parameters additively improved diagnostic accuracy for breast cancer.

Pectoral muscle deletion on a mammogram to aid in the early diagnosis of breast cancer

Mammograms are the images used by radiologists to diagnose breast cancer. In this diagnosis, the pectoral muscle appears in mammograms in oblique mediolateral views (MLO) of the right breast and another in the left breast appears in cranio-caudal views which are marked with (CC). Considering that the pectoral muscle has the same density as the small, suspicious masses in the image, its presence in the image being processed could also require detection procedures. In this paper, we present a new general framework for pectoral muscle suppression which is the first work in the analysis of a mammography image. As a result, we proceed to four stages of image processing. The first step is to orient the image if necessary, then use a pre-processing which is to enhance the contrast of the image, and remove the digital lines of the image by morphological filters, apply a filter median. The third step involves segmenting all of the pectoral muscles, which involves threshold the entire image. The final step is to perform a pectoral muscle removal according to the orientation of the muscle in the image, which will be based on the development of the Hough transform for the recognition of borderline detections of the pectoral muscle. Some results obtained on the different images are discussed and compared with other methods (risk assessments). Evaluation of our method shows a significant improvement in performance in removing the pectoral muscle.&#x0D; Keywords: Breast cancer, Mammogram, Pectoral muscle, Hough transform.

Outcome of mammography examination among asymptomatic women.

e13558 Background: Early detection of breast cancer is important in reducing mortality, morbidity and high socio-economic burden associated with it. Mammography is currently the primary imaging modality used as a screening tool to detect early breast cancer in women experiencing no symptoms as they are most curable in the early stage. The aim of the study is to determine the mammographic outcome in asymptomatic women who presented for mammographic breast examination in Abuja,Nigeria. Methods: This descriptive cross-sectional study comprises of 113 asymptomatic women who presented for mammographic examination at the Radiology department of University of Abuja Teaching Hospital, Gwagwalada from March 2015 to December 2018. Two basic views (craniocaudal and mediolateral views) of the breast were obtained using EXR-650 machine.Additional views were obtained when necessary. Images of the breast were review by radiologist. Results: The mean age of study population was 40.72 ±10.45years with age range of 35 and 65 years. The mammographic outcome among asymptomatic women who had mammographic examination was negative in 69(61.1%) women and positive in 44(38.9%). The differences observed between the positive and negative mammographic outcome was statistically significant p = 0.01. The positive outcome noted in mammograms of women examined were: benign mass in 18(15.9%) women; 9(8.0%) had benign calcification; 7(6.2%) showed architectural distortion; 5(4.4%) was inconclusive; focal asymmetry in 3(2.6%); and suspicious mass in 2(1.8%). Conclusions: Mammogram is an important tool for screening and diagnoses of breast pathologies. In this study, screening of women reveals various benign and malignant breast changes which necessitate early interventions. Early detection of breast cancer save lives. [Table: see text]

Multi-stage System for Preprocessing Mammograms

Breast image analysis is important for the detection and diagnosis of various diseases, but more specifically, breast cancer. The most popular and widely used screening tool is mammography. It gives a better result than other screening tools for breast image analysis. Mammograms are difficult to analyze because of noise, poor quality, artifacts, and the inclusion of pectoral muscle. The technique proposed in this paper has several stages for preprocessing mammograms. First, artifacts and the background are removed by changing the original image into a binary image using Otsu’s thresholding. The medio-lateral view of mammography images has two orientations: left or right. Orientation detection in the next step is important for pectoral muscle removal, which is done using the line segment method. Contrast is enhanced using contrast limited adaptive histogram equalization, and noise is reduced using a median filter. The proposed technique was applied to a publicly available dataset from the Mammographic Image Analysis Society, from which 322 images were analyzed, and performance parameters calculated. After calculations for accuracy, sensitivity, and specificity, the accuracy achieved was 94.62%, sensitivity achieved was 90.21%, and specificity was 97.20%.

Subsequent Marking under Ultrasound Guidance of Vacuum-Assisted Breast Biopsy Areas after Receipt of Histology: A Feasibility Study of a New Technique

Purpose: The aim of this study was to evaluate the feasibility and the accuracy of a secondary, metachronous ultrasound (US)-guided marking of the stereotactic vacuum-assisted breast biopsy (ST-VABB) area. Materials and Methods: The institutional ethics committee approved the study. The retrospective study included 98 patients. In ST-VABB of 45 women, no tissue markers were deployed at the biopsy site, even if no residual calcifications remained. After histology proved the necessity for a subsequent operation, the biopsy site was marked under US guidance using a coil marker. All interventions were technically successful. No complications occurred. Mammography was done to visualize the coil deployment. The distances from the center of the lesion and the biopsy cavity to the coil location were measured in both planes to evaluate the accuracy of the marking procedure. Results: In 24 of the 46 cases, the whole lesion was biopsied without residual elements. The mean time between ST-VABB and sonographic marking of the lesion was 9.7 days (median 6.5). The biopsy cavity could be detected in 40 (87%) cases and thus marked exactly. The mean time of US-guided marking was 12.5 min. The mean distance between the coil and the target lesion was 0.6 ± 1.5 cm in the craniocaudal (cc) view and 0.5 ± 1.5 cm in the mediolateral (ml) view for all markings. The mean delta value from the distance nipple–original lesion and from the distance nipple–coil was 0.85 ± 1.2 cm (median 0.5) in the cc view and 0.88 ± 1.2 cm (median 0.6) in the ml view for all cases. Clip migration was not observed. Conclusion: Our study demonstrates the feasibility and the technical success of secondary metachronous coil marking of the biopsy site under US guidance after receipt of histology. This approach seems to be a cost-effective alternative to the standard procedure of the primary coil marking especially in all completely removed lesions. It may offer advantages for allergic patients.

Abstract PD9-03: A combination of intra- and peri-lesional deep learning classifiers from multiple views enables accurate diagnosis of architectural distortion malignancy with digital breast tomosynthesis

Abstract Background: 3D digital breast tomosynthesis (DBT) offers greater sensitivity than 2D mammography to the presence of architectural distortions (AD), a suspicious finding requiring biopsy. This higher sensitivity to AD is only beneficial if DBT can differentiate benign AD, such as radial scars (RS) from malignant AD (MAD). Automated analysis of clinical images using a neural network has potential to enhance the accuracy of breast cancer diagnosis, but such applications focus on screening and currently struggle with more challenging diagnoses (e.g. MAD vs. RS) where computer assistance would provide great clinical utility. In this work, we define a deep learning (DL) framework utilizing amalgamate predictions from a group of neural networks, specialized for multiple lesion regions and different acquisitional views to distinguish MAD from RS. Methods: A retrospective analysis was conducted of 69 patients screened at a single institution, each with AD visible on one or more DBT views. Stereotactic core biopsy determined 42 AD to be MAD and the remaining 27 to be RS. An attending breast radiologist manually identified a circular region of interest (ROI) containing the AD on a single slice of the mediolateral (ML, n=72) and/or craniocaudal (CC, n=68) views. 56 patients were utilized for training and tuning the DL networks in cross-validation, while 25 volumes from 13 patients (8 MAD, 5 RS) were held for independent testing. Volumetric samples of size 24x24x12 pixels were extracted randomly within each ROI for input to the networks. Two neural networks were trained to separately recognize patterns of MAD at the intra- and peri-lesional region. Pairs of networks were trained separately for analysis of the CC and the MLO view, resulting in 4 possible network variations. These networks were first assessed individually by cross-validation in the training set, then predictions from each were combined for diagnosis in the independent testing set. Performance was assessed by area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and accuracy. Results: All networks individually achieved strong performance in distinguishing RS from MAD when assessed via cross-validation in the training set. On the independent test set, networks trained to recognize patterns of malignancy at the center of the distortion outperformed those focused on the lesion margin. The CC view was found to be most discriminating in neural network assessment. Independent Test PerformanceNetwork TypeAUCSensitivitySpecificityIntralesional, CC0.650.800.50Perilesional, CC0.510.400.63Intralesional, ML0.490.600.38Perilesional, ML0.550.600.50Combined Performance0.800.601.0 When predictions from these networks were combined for diagnosis on the independent validation set, performance increased significantly to an AUC of 0.8, with a sensitivity of 0.6 and a specificity of 1.0. Conclusions: In this investigation, we demonstrated that specialized deep learning classifiers, created to address the diagnosis of MAD vs RS, can assist in diagnostic settings without generating false negatives. Our approach combines neural networks to make a diagnosis based on various spatial regions from multiple DBT views. Our approach demonstrated very high specificity which may reduce the number of benign RS biopsies and surgical excisions for AD detected by DBT. Future work will further validate these findings on a larger, multi-institutional cohort and explore their influence on clinical decision-making. Citation Format: Tristan Maidment, Nathaniel Braman, Yijiang Chen, Farhad Mehrkhani, Uliyana Yankevich, Donna Plecha, Anant Madabhushi. A combination of intra- and peri-lesional deep learning classifiers from multiple views enables accurate diagnosis of architectural distortion malignancy with digital breast tomosynthesis [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr PD9-03.

Accuracy of Distinguishing Atypical Ductal Hyperplasia From Ductal Carcinoma In Situ With Convolutional Neural Network-Based Machine Learning Approach Using Mammographic Image Data.

OBJECTIVE. The purpose of this study was to test the hypothesis that convolutional neural networks can be used to predict which patients with pure atypical ductal hyperplasia (ADH) may be safely monitored rather than undergo surgery. MATERIALS AND METHODS. A total of 298 unique images from 149 patients were used for our convolutional neural network algorithm. A total of 134 images from 67 patients with ADH that had been diagnosed by stereotactic-guided biopsy of calcifications but had not been upgraded to ductal carcinoma in situ or invasive cancer at the time of surgical excision. A total of 164 images from 82 patients with mammographic calcifications indicated that ductal carcinoma in situ was the final diagnosis. Two standard mammographic magnification views of the calcifications (a craniocaudal view and a mediolateral or lateromedial view) were used for analysis. Calcifications were segmented using an open-source software platform and images were resized to fit a bounding box of 128 × 128 pixels. A topology with 15 hidden layers was used to implement the convolutional neural network. The network architecture contained five residual layers and dropout of 0.25 after each convolution. Patients were randomly separated into a training-and-validation set (80% of patients) and a test set (20% of patients). Code was implemented using open-source software on a workstation with an open-source operating system and a graphics card. RESULTS. The AUC value was 0.86 (95% CI, ± 0.03) for the test set. Aggregate sensitivity and specificity were 84.6% (95% CI, ± 4.0%) and 88.2% (95% CI, ± 3.0%), respectively. Diagnostic accuracy was 86.7% (95% CI, ± 2.9). CONCLUSION. It is feasible to apply convolutional neural networks to distinguish pure atypical ductal hyperplasia from ductal carcinoma in situ with the use of mammographic images. A larger dataset will likely result in further improvement of our prediction model.

Synthesizing mammogram from digital breast tomosynthesis

The purpose of this study is to develop a new method for generating synthesized mammogram (SM) from digital breast tomosynthesis (DBT) and to assess its potential as an adjunct to DBT. We first applied multiscale bilateral filtering to the reconstructed DBT slices to enhance the high-frequency features and reduce noise. A maximum intensity projection (MIP) image was then obtained from the high-frequency components of the DBT slices. A multiscale image fusion method was designed to combine the MIP image and the central DBT projection view into an SM and further enhance the high-frequency features. We conducted a pilot reader study to visually assess the image quality of SM in comparison to full field digital mammograms (FFDM). For each DBT craniocaudal or mediolateral view, a clinical FFDM of the corresponding view was retrospectively collected. Three MQSA radiologists, blinded to the pathological and other clinical information, independently interpreted the SM and the corresponding FFDM side by side marked with the lesion locations. The differences in the BI-RADS assessments of both MCs and masses between SM and FFDM did not achieve statistical significance for all three readers. The conspicuity of MCs on SM was superior to that on FFDM and the BI-RADS assessments of MCs were comparable while the conspicuity of masses on SM was degraded and interpretation on SM was less accurate than that on FFDM. The SM may be useful for efficient prescreening of MCs in DBT but the DBT should be used for detection and characterization of masses.

Minimization of annotation work: diagnosis of mammographic masses via active learning

The prerequisite for establishing an effective prediction system for mammographic diagnosis is the annotation of each mammographic image. The manual annotation work is time-consuming and laborious, which becomes a great hindrance for researchers. In this article, we propose a novel active learning algorithm that can adequately address this problem, leading to the minimization of the labeling costs on the premise of guaranteed performance.Our proposed method is different from the existing active learning methods designed for the general problem as it is specifically designed for mammographic images. Through its modified discriminant functions and improved sample query criteria, the proposed method can fully utilize the pairing of mammographic images and select the most valuable images from both the mediolateral and craniocaudal views. Moreover, in order to extend active learning to the ordinal regression problem, which has no precedent in existing studies, but is essential for mammographic diagnosis (mammographic diagnosis is not only a classification task, but also an ordinal regression task for predicting an ordinal variable, viz. the malignancy risk of lesions), multiple sample query criteria need to be taken into consideration simultaneously. We formulate it as a criteria integration problem and further present an algorithm based on self-adaptive weighted rank aggregation to achieve a good solution.The efficacy of the proposed method was demonstrated on thousands of mammographic images from the digital database for screening mammography. The labeling costs of obtaining optimal performance in the classification and ordinal regression task respectively fell to 33.8 and 19.8 percent of their original costs. The proposed method also generated 1228 wins, 369 ties and 47 losses for the classification task, and 1933 wins, 258 ties and 185 losses for the ordinal regression task compared to the other state-of-the-art active learning algorithms.By taking the particularities of mammographic images, the proposed AL method can indeed reduce the manual annotation work to a great extent without sacrificing the performance of the prediction system for mammographic diagnosis.

Radiological and osteological study of the pelvic limbs in free-ranging capybaras (Hydrochoerus hydrochaeris).

This study aimed to describe the radiographic anatomy and osteology and to evaluate angular radiographic measurements-Norberg angle, inclination angle and anteversion angle-of the pelvic limbs in free-ranging capybaras. Twenty cadavers of free-ranging capybaras (Hydrochoerus hydrochaeris), including five adults and 15 subadults, were studied. Ventrodorsal, craniocaudal, dorsoplantar and mediolateral radiographic views of the pelvic limbs were obtained. The radiographic features were described together with bone samples. The hip bone (os coxae), shaped like an isosceles trapezoid, was elongated and narrow with the presence of an oblong obturator foramen, sagittal ilial wing and rectilinear ilial body. The femoral shaft was relatively straight, while the greater trochanter was projected above the femoral head. No sesamoid bones of the gastrocnemius and popliteus muscles were observed radiographically or for those animals used in gross macroscopy. The fibula was located lateral and parallel to the tibia. Eight tarsal bones, four metatarsal bones and three digits were identified. The mean radiographic measurements included Norberg angle of 125.9°; respective angles of femoral inclination by the Hauptman B and Tomlinson methods of 139.9 and 141°; anteversion angle of the femoral head and neck of 29.80°. The bones of the pelvic limbs in capybaras have several anatomical characters and radiological features that are shared with members of the caviomorph superfamily Cavioidea. The radiographic angles measured in this study help characterize the functional morphology of this species.