The Importance of Imaging Informatics and Informaticists in the Implementation of AI Publication date: Available online 19 October 2019 Source: Academic Radiology Author(s): Tessa S. Cook Imaging informatics is critical to the success of AI implementation in radiology. An imaging informaticist is a unique individual who sits at the intersection of clinical radiology, data science, and information technology. With the ability to understand each of the different domains and translate between the experts in these domains, imaging informaticists are now essential players in the development, evaluation, and deployment of AI in the clinical environment.

Artificial Intelligence in Medicine: Where Are We Now? Publication date: Available online 19 October 2019 Source: Academic Radiology Author(s): Sagar Kulkarni, Nuran Seneviratne, Mirza Shaheer Baig, Ameer Hamid Ahmed Khan Artificial intelligence in medicine has made dramatic progress in recent years. However, much of this progress is seemingly scattered, lacking a cohesive structure for the discerning observer. In this article, we will provide an up-to-date review of artificial intelligence in medicine, with a specific focus on its application to radiology, pathology, ophthalmology, and dermatology. We will discuss a range of selected papers that illustrate the potential uses of artificial intelligence in a technologically advanced future.

Collaborative Learning in Radiology: From Peer Review to Peer Learning and Peer Coaching Publication date: Available online 18 October 2019 Source: Academic Radiology Author(s): Alison L. Chetlen, Jonelle Petscavage-Thomas, Rekha A. Cherian, Adam Ulano, Sadhna B. Nandwana, Nicole E. Curci, Robert T. Swanson, Rick Artrip, Tharakeswara K Bathala, Lori Mankowski Gettle, L. Alexandre Frigini Background A Radiology Research Alliance Task Force was assembled in 2018 to review the literature on peer review and report on best practices for peer learning and peer coaching. Findings This report provides a historical perspective on peer review and the transition to peer collaborative learning and peer coaching. Most forms of current peer review have fulfilled regulatory requirements but have failed to significantly impact quality improvement or learning opportunities. Peer learning involves joint intellectual efforts by two or more individuals to study best practices and review error collaboratively. Peer coaching is a process in which individuals in a trusted environment work to expand, refine, and build new skills in order to facilitate self-directed learning and professional growth. We discuss the value in creating opportunities for peer learning and peer coaching. Conclusion Peer collaborative learning combined with peer coaching provides opportunities for teams to learn and grow together, benefit from each other's expertise and experience, improve faculty morale, and provide more opportunities for collaborations between faculty.

The Diagnostic Value of MRI for Preoperative Staging in Patients with Endometrial Cancer: A Meta-Analysis Publication date: Available online 18 October 2019 Source: Academic Radiology Author(s): Qiu Bi, Yuhui Chen, Kunhua Wu, Junna Wang, Ying Zhao, Bo Wang, Ji Du Objectives To assess the diagnostic accuracy of magnetic resonance imaging (MRI) for detecting myometrial invasion, cervical invasion, and lymph node metastases in endometrial cancer. Materials and Methods A systematic literature search was performed in PubMed, Embase, Cochrane Library, Web of Science, and Clinical trials. The methodological quality of each study was assessed by using the standard Quality Assessment of Diagnostic Accuracy Studies-2. Statistical analysis included evaluating publication bias, assessing threshold effect, exploring heterogeneity, pooling data, meta-regression, forest plot, and summary receiver-operating characteristics curves construction. Results Fourteen studies could be analyzed. For detecting deep myometrial invasion, the pooled sensitivity and specificity were 0.79 and 0.81 respectively, and patients younger than 60 years old demonstrated higher sensitivity (0.84) and specificity (0.90). The diagnostic accuracy is highest by jointly using T2-weighted image, dynamic contrast-enhanced MRI, and diffusion weighted imaging to detect the deep myometrial invasion. There were low sensitivity and high specificity for the diagnosis of cervical invasion (0.53, 0.95), cervical stromal invasion (0.50, 0.95), pelvic or/and para-aortic lymph node metastases (0.59, 0.95), and pelvic lymph node metastases (0.65, 0.95). Conclusion MRI has good diagnostic performance for assessing myometrial invasion in patients with endometrial cancer, especially in patients younger than 60 years old. Dynamic contrast-enhanced MRI and diffusion weighted imaging can help improve sensitivity and specificity for detecting myometrial invasion. MRI shows high specificity for detecting cervical invasion and lymph node metastases in endometrial cancer.

Benefits of Silent DWI MRI in Success Rate, Image Quality, and the Need for Secondary Sedation During Brain Imaging of Children of 3–36 Months of Age Publication date: Available online 17 October 2019 Source: Academic Radiology Author(s): Xi Zhu, Jing Ye, Zhuqing Bao, Xianfu Luo, Qingqiang Zhu, Songan Shang, Weiqiang Dou, Wei Xia Rationale and Objectives Silent T1W and T2W magnetic resonance imaging (MRI) can be used to study myelination in children, but the success rate of silent diffusion-weighted imaging is unknown. This study aimed to evaluate the success rate and image quality of silent MRI for the brain of children. Materials and Methods This was a retrospective study of 3–36-month children who underwent silent or conventional brain MRI at the People's Hospital of Northern Jiangsu from 01/2015 to 02/2018. The success rates were compared. The acoustic noise of each sequence was measured using a decibel meter. The signal-to-noise ratio and contrast-to-noise ratio of the diffusion-weighted imaging, T2W, and T1W sequences were analyzed. Subjective image quality (lesion delineation, visibility, gray-white differentiation, and overall diagnostic usefulness) was determined. Results The success rate of silent MRI (n = 443) was higher than that of conventional MRI (n = 391) (97.7% vs. 88.2%, p < 0.001). The acoustic noise of all silent sequences was lower than that of the conventional sequence (all p < 0.05). Silent sequences showed decreased signal-to-noise ratio vs. conventional sequences but increased contrast-to-noise ratio (all p < 0.05). Lesion delineation was not significantly different. Lesion visibility and gray-white differentiation of all silent sequences were higher (all p < 0.05). The overall diagnostic usefulness of the silent group was higher (p < 0.001). Conclusion Silent MRI can effectively improve the success rate of MRI in children of 3–36 months. Noise is reduced, and the overall diagnostic usefulness is higher than that of conventional MRI. Silent MRI is more suitable for children's brain scan than conventional MRI.

Safety and Efficacy Studies of Vertebroplasty with Dual Injections for the Treatment of Osteoporotic Vertebral Compression Fractures: Preliminary Report Publication date: Available online 16 October 2019 Source: Academic Radiology Author(s): Pijian Cao, Weimin Hao, Lu Zhang, Qinglin Zhang, Xunwei Liu, Min Li Purpose To evaluate the clinical safety and efficacies of percutaneous vertebroplasty (PVP), percutaneous vertebroplasty with dual injections (PVPDI), and percutaneous kyphoplasty (PKP) for the treatment of osteoporotic vertebral compression fractures (OVCFs), a retrospective study of 90 patients with OVCFs who had been treated by PVP (n = 30), PVPDI (n = 30), and PKP (n = 30) was conducted in this work. Methods The clinical efficacies of these three treatments were evaluated by comparing their PMMA cement leakages, cement patterns, height restoration percentages, wedge angles, visual analogue scales, and Oswestry disability index (ODI) at the pre- and postoperative time points. Results Ten percent, 6.7%, and 0% of patients had PMMA leakage in PVP, PVPDI, and PKP groups, respectively. Three (solid, trabecular, and mixed patterns), two (trabecular and mixed patterns), and two (solid and mixed patterns) types of cement patterns were observed in PVP, PVPDI, and PKP groups, respectively. PVP and PVPDI treatments had similar and less height restoration ability than PKP treatment. All the PVP, PVPDI, and PKP treatments had significant and similar ability in pain relief and functional recovery ability for the treatment of OVCFs. Microfractures after the surgery occurred after PVP and PKP treatments. Conclusion These results indicate minimally invasive techniques were effective methods for the treatment of OVCFs. Moreover, these initial outcomes suggest PVPDI treatment has great value and is worth promoting vigorously in orthopedics clinics.

Burnout Phenomenon and Its Predictors in Radiology Residents Publication date: Available online 16 October 2019 Source: Academic Radiology Author(s): Abdulmajeed Bin Dahmash, Fawziah Khalid Alorfi, Abdulaziz Alharbi, Abdulrahman Aldayel, Ahmed M. Kamel, Mohammed Almoaiqel Rationale and Objectives The purpose of this study is to evaluate the prevalence of burnout and its associated risk factors in radiology residents in Saudi Arabia. Materials and Methods This cross-sectional study was conducted in February 2019, and all radiology residents in Riyadh, Saudi Arabia, were invited to complete a survey that contained a validated measure of burnout (Maslach Burnout Inventory-Human Services Survey) alongside possible predictors of burnout. Results A total of 108 responses were received, for a response rate of 49.7%. High overall burnout was reported by 24.1% of respondents, high emotional exhaustion (EE) by 56.5%, high depersonalization by 31.5%, and low sense of personal accomplishment (PA) by 64.8%. The significant predictors of burnout included satisfaction with work/life balance (OR = 0.35, 95% CI = 0.03 to 0.43, p = 0.002) and exercising (OR = 0.31, 95% CI = 0.1 to 1, p = 0.07). Married residents were more prone to have a low sense of PA in addition to dissatisfied residents with hospital staff appreciation (OR = 4.8, 95% CI = 1.48 to 15.5, p = 0.01) and (OR = 0.59, 95% CI = 0.37 to 0.94, p = 0.03), respectively. Conclusion One-fourth of the radiology residents studied showed high rates of burnout, and more than half the residents reported high rates of EE. The residents scored very poorly in the sense of PA. The radiology residents who were satisfied with their work/life balance had lower burnout rates, in addition to lower EE and a higher sense of PA.

The Value of Contrast-Enhanced CT in the Detection of Residual Disease After Neo-Adjuvant Chemotherapy in Ovarian Cancer Publication date: Available online 16 October 2019 Source: Academic Radiology Author(s): He An, Keith W.H. Chiu, K.Y. Tse, Hextan Y.S. Ngan, Pek-Lan Khong, Elaine Y.P. Lee Rationale and Objectives To evaluate the diagnostic performance of contrast-enhanced computed tomography (CT) in predicting residual disease following neo-adjuvant chemotherapy (NACT) in stage III/IV ovarian cancer. Materials and Methods This was a retrospective observational cohort study including consecutive patients with primary stage III/IV ovarian cancer who received NACT before interval debulking surgery. CT findings before interval debulking surgerywere correlated with histological/surgical findings. Diagnostic characteristics were calculated on patient-based and lesion-based analyses. False negative results on peritoneal carcinomatosis detection were correlated with lesion size and site. Results On patient-based analysis, CT (n = 58) had a sensitivity, specificity, positive predictive value, negative predictive value and accuracy of 92.16%, 57.14%, 94.00%, 50.00%, and 87.93%. On lesion-based analysis, the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 63.01%, 73.47%, 82.51%, 50.00%, and 66.51%. False negative results were associated with lesion size (p < 0.001). The diagnostic performance of CT on the detection of peritoneal carcinomatosis was low at the subdiaphragmatic spaces, bowel serosa and mesentery (p < 0.001). Conclusion CT had low negative predictive value in determining residual disease following NACT on both patient-based and lesion-based analyses, especially for non-measurable lesions and at the subdiaphragmatic spaces, bowel serosa and mesentery.

Initial Clinical Experience of Virtual Monoenergetic Imaging Improves Stent Visualization in Lower Extremity Run-Off CT Angiography by Dual-Layer Spectral Detector CT Publication date: Available online 15 October 2019 Source: Academic Radiology Author(s): Daming Zhang, Yanting Xie, Yining Wang, Ning Guo, Yun Wang, Zhengyu Jin, Huadan Xue Rationale and Objectives Virtual monoenergetic imaging (VMI) may improve stent visualization in lower extremity run-off computed tomography angiography. The purpose of this study was to evaluate the image quality (IQ) of stents and to determine the optimal kiloelectron volt (keV) level of VMI images for stent evaluation compared to conventional CT images. Materials and Methods This study included 32 patients with prior stent placement who underwent run-off computed tomography angiography on a dual-layer spectral detector CT scanner. Thirteen image series were evaluated for each stent, including conventional CT and 12 VMI datasets from 40 keV to 150 keV obtained in 10-keV intervals. Attenuation, SD, contrast-to-noise ratio, and signal-to-noise ratio of the native vessel and the vessel with a stent were evaluated. The diameter of the stent was measured in all 13 image series. The IQ was evaluated by two readers using a five-point scale (1 = poor IQ, 5 = excellent IQ). Results A total of 39 stents in 29 patients were evaluated. Compared to conventional CT, attenuation of the native vessel and the vessel with a stent was higher at 40–60 keV, and the SD was equal or lower at 50–150 keV. Based on the attenuation and SD of VMI images, the contrast-to-noise ratio and signal-to-noise ratio were higher at 40–70 keV, among which the highest ratios were obtained at 40 keV. The stent diameter was equal or larger at 60–150 keV, and the lowest stent diameter underestimation occurred at 100 keV. The IQ was equal or higher, ranging from 60 to 100 keV in comparison with conventional CT, and the highest IQ score occurred at 90 keV. Conclusion This quantitative and qualitative assessment of VMI images and conventional images indicated that IQ improvement and more accurate stent lumen evaluation on lower extremity run-off CT angiography can be achieved by dual-layer spectral detector CT.

Unboxing AI - Radiological Insights Into a Deep Neural Network for Lung Nodule Characterization Publication date: Available online 14 October 2019 Source: Academic Radiology Author(s): Vasantha Kumar Venugopal, Kiran Vaidhya, Murali Murugavel, Abhijith Chunduru, Vidur Mahajan, Suthirth Vaidya, Digvijay Mahra, Akshay Rangasai, Harsh Mahajan Rationale and Objectives To explain predictions of a deep residual convolutional network for characterization of lung nodule by analyzing heat maps. Materials and Methods A 20-layer deep residual CNN was trained on 1245 Chest CTs from National Lung Screening Trial (NLST) trial to predict the malignancy risk of a nodule. We used occlusion to systematically block regions of a nodule and map drops in malignancy risk score to generate clinical attribution heatmaps on 103 nodules from Lung Image Database Consortium image collection and Image Database Resource Initiative (LIDC-IDRI) dataset, which were analyzed by a thoracic radiologist. The features were described as heat inside nodule -bright areas inside nodule, peripheral heat continuous/interrupted bright areas along nodule contours, heat in adjacent plane -brightness in scan planes juxtaposed with the nodule, satellite heat - a smaller bright spot in proximity to nodule in the same scan plane, heat map larger than nodule bright areas corresponding to the shape of the nodule seen outside the nodule margins and heat in calcification. Results These six features were assigned binary values. This feature vector was fedinto a standard J48 decision tree with 10-fold cross-validation, which gave an 85 % weighted classification accuracy with a 77.8% True Positive (TP) rate, 8% False Positive (FP) rate for benign cases and 91.8% TP and 22.2% FP rates for malignant cases. Heat Inside nodule was more frequently observed in nodules classified as malignant whereas peripheral heat, heat in adjacent plane, and satellite heat were more commonly seen in nodules classified as benign. Conclusion We discuss the potential ability of a radiologist to visually parse the deep learning algorithm generated “heat map” to identify features aiding classification.