Artificial Intelligence with Statistical Confidence Scores for Detection of Acute or Subacute Hemorrhage on Noncontrast CT Head Scans.

Purpose: To present a method that automatically detects, subtypes, and locates acute or subacute intracranial hemorrhage (ICH) on noncontrast CT (NCCT) head scans; generates detection confidence scores to identify high-confidence data subsets with higher accuracy; and improves radiology worklist prioritization. Such scores may enable clinicians to better use artificial intelligence (AI) tools. Materials and Methods: This retrospective study included 46 057 studies from seven "internal" centers for development (training, architecture selection, hyperparameter tuning, and operating-point calibration; n = 25 946) and evaluation (n = 2947) and three "external" centers for calibration (n = 400) and evaluation (n = 16 764). Internal centers contributed developmental data, whereas external centers did not. Deep neural networks predicted the presence of ICH and subtypes (intraparenchymal, intraventricular, subarachnoid, subdural, and/or epidural hemorrhage) and segmentations per case. Two ICH confidence scores are discussed: a calibrated classifier entropy score and a Dempster-Shafer score. Evaluation was completed by using receiver operating characteristic curve analysis and report turnaround time (RTAT) modeling on the evaluation set and on confidence score-defined subsets using bootstrapping. Results: The areas under the receiver operating characteristic curve for ICH were 0.97 (0.97, 0.98) and 0.95 (0.94, 0.95) on internal and external center data, respectively. On 80% of the data stratified by calibrated classifier and Dempster-Shafer scores, the system improved the Youden indexes, increasing them from 0.84 to 0.93 (calibrated classifier) and from 0.84 to 0.92 (Dempster-Shafer) for internal centers and increasing them from 0.78 to 0.88 (calibrated classifier) and from 0.78 to 0.89 (Dempster-Shafer) for external centers (P < .001). Models estimated shorter RTAT for AI-prioritized worklists with confidence measures than for AI-prioritized worklists without confidence measures, shortening RTAT by 27% (calibrated classifier) and 27% (Dempster-Shafer) for internal centers and shortening RTAT by 25% (calibrated classifier) and 27% (Dempster-Shafer) for external centers (P < .001). Conclusion: AI that provided statistical confidence measures for ICH detection on NCCT scans reliably detected and subtyped hemorrhages, identified high-confidence predictions, and improved worklist prioritization in simulation.Keywords: CT, Head/Neck, Hemorrhage, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2022.

Semi-automated delineation of breast cancer tumors and subsequent materialization using three-dimensional printing (rapid prototyping).

OBJECTIVE: Three-dimensional (3D) printing has become widely available, and a few cases of its use in clinical practice have been described. The aim of this study was to explore facilities for the semi-automated delineation of breast cancer tumors and to assess the feasibility of 3D printing of breast cancer tumors. METHODS: In a case series of five patients, different 3D imaging methods-magnetic resonance imaging (MRI), digital breast tomosynthesis (DBT), and 3D ultrasound-were used to capture 3D data for breast cancer tumors. The volumes of the breast tumors were calculated to assess the comparability of the breast tumor models, and the MRI information was used to render models on a commercially available 3D printer to materialize the tumors. RESULTS: The tumor volumes calculated from the different 3D methods appeared to be comparable. Tumor models with volumes between 325 mm3 and 7,770 mm3 were printed and compared with the models rendered from MRI. The materialization of the tumors reflected the computer models of them. CONCLUSION: 3D printing (rapid prototyping) appears to be feasible. Scenarios for the clinical use of the technology might include presenting the model to the surgeon to provide a better understanding of the tumor's spatial characteristics in the breast, in order to improve decision-making in relation to neoadjuvant chemotherapy or surgical approaches. J. Surg. Oncol. 2017;115:238-242. © 2016 Wiley Periodicals, Inc.

Performance of Photon-Counting Breast Computed Tomography, Digital Mammography, and Digital Breast Tomosynthesis in Evaluating Breast Specimens.

RATIONALE AND OBJECTIVES: This study compared a novel photon-counting breast computed tomography (pcBCT) system with digital mammography (DM) and digital breast tomosynthesis (DBT) systems. For this reason, surgical specimens were examined with all three techniques and rated by three observers. MATERIALS AND METHODS: A total of 30 surgical specimens were investigated with DM, DBT, and pcBCT; the associated images were shown to three experienced radiologists. Findings (22 microcalcifications and 23 mass lesions) were recorded and compared to the results of the pathological examination. Sensitivity and specificity for detection of microcalcifications and lesions were calculated and displayed using receiver operating characteristic curves. RESULTS: Sensitivity for microcalcifications was 82% for DM, 70% for DBT, and 85% for pcBCT. Specificity for microcalcifications was 71% for DM, 75% for DBT, and 83% for pcBCT. Sensitivity for lesions was 45% for DM, 62% for DBT, and 65% for pcBCT. Specificity for lesions was 76% for DM, 62% for DBT, and 76% for pcBCT. CONCLUSIONS: pcBCT showed a comparable or superior performance compared to the clinically approved DM and DBT systems. Mass lesion detectability can be increased further by the use of contrast media.

Mammographic density is the main correlate of tumors detected on ultrasound but not on mammography.

Although mammography screening programs do not include ultrasound examinations, some diagnostic units do provide women with both mammography and ultrasonography. This article is concerned with estimating the risk of a breast cancer patient diagnosed in a hospital-based mammography unit having a tumor that is visible on ultrasound but not on mammography. A total of 1,399 women with invasive breast cancer from a hospital-based diagnostic mammography unit were included in this retrospective study. For inclusion, mammograms from the time of the primary diagnosis had to be available for computer-assisted assessment of percentage mammographic density (PMD), as well as Breast Imaging Reporting and Data System (BIRADS) assessment of mammography. In addition, ultrasound findings were available for the complete cohort as part of routine diagnostic procedures, regardless of any patient or imaging characteristics. Logistic regression analyses were conducted to identify predictors of mammography failure, defined as BIRADS assessment 1 or 2. The probability that the visibility of a tumor might be masked at diagnosis was estimated using a regression model with the identified predictors. Tumors were only visible on ultrasound in 107 cases (7.6%). PMD was the strongest predictor for mammography failure, but age, body mass index and previous breast surgery also influenced the risk, independently of the PMD. Risk probabilities ranged from 1% for a defined low-risk group up to 40% for a high-risk group. These findings might help identify women who should be offered ultrasound examinations in addition to mammography.

A hybrid method towards automated nipple detection in 3D breast ultrasound images.

In clinical work-up of breast cancer, nipple position is an important marker to locate lesions. Moreover, it serves as an effective landmark to register a 3D automated breast ultrasound (ABUS) images to other imaging modalities, e.g., X-ray mammography, tomosynthesis or magnetic resonance imaging (MRI). However, the presence of speckle noises caused by the interference waves and variant imaging directions poses challenges to automatically identify nipple positions. In this work, a hybrid fully automatic method to detect nipple positions in ABUS images is presented. The method extends the multi-scale Laplacian-based method that we proposed previously, by integrating a specially designed Hessian-based method to locate the shadow area beneath the nipple and areola. Subsequently, the likelihood maps of nipple positions generated by both methods are combined to build a joint-likelihood map, where the final nipple position is extracted. To validate the efficiency and robustness, the extended hybrid method was tested on 926 ABUS images, resulting in a distance error of 7.08±10.96 mm (mean±standard deviation).