Thomas W. Loehfelm

A Cloud-based Multimodality Case File for Mobile Devices

Recent improvements in Web and mobile technology, along with the widespread use of handheld devices in radiology education, provide unique opportunities for creating scalable, universally accessible, portable image-rich radiology case files. A cloud database and a Web-based application for radiologic images were developed to create a mobile case file with reasonable usability, download performance, and image quality for teaching purposes. A total of 75 radiology cases related to breast, thoracic, gastrointestinal, musculoskeletal, and neuroimaging subspecialties were included in the database. Breast imaging cases are the focus of this article, as they best demonstrate handheld display capabilities across a wide variety of modalities. This case subset also illustrates methods for adapting radiologic content to cloud platforms and mobile devices. Readers will gain practical knowledge about storage and retrieval of cloud-based imaging data, an awareness of techniques used to adapt scrollable and high-resolution imaging content for the Web, and an appreciation for optimizing images for handheld devices. The evaluation of this software demonstrates the feasibility of adapting images from most imaging modalities to mobile devices, even in cases of full-field digital mammograms, where high resolution is required to represent subtle pathologic features. The cloud platform allows cases to be added and modified in real time by using only a standard Web browser with no application-specific software. Challenges remain in developing efficient ways to generate, modify, and upload radiologic and supplementary teaching content to this cloud-based platform. Online supplemental material is available for this article.

Digitizing an Analog Radiography Teaching File Under Time Constraint: Trade-Offs in Efficiency and Image Quality

We digitized the radiography teaching file at Black Lion Hospital (Addis Ababa, Ethiopia) during a recent trip, using a standard digital camera and a fluorescent light box. Our goal was to photograph every radiograph in the existing library while optimizing the final image size to the maximum resolution of a high quality tablet computer, preserving the contrast resolution of the radiographs, and minimizing total library file size. A secondary important goal was to minimize the cost and time required to take and process the images. Three workers were able to efficiently remove the radiographs from their storage folders, hang them on the light box, operate the camera, catalog the image, and repack the radiographs back to the storage folder. Zoom, focal length, and film speed were fixed, while aperture and shutter speed were manually adjusted for each image, allowing for efficiency and flexibility in image acquisition. Keeping zoom and focal length fixed, which kept the view box at the same relative position in all of the images acquired during a single photography session, allowed unused space to be batch-cropped, saving considerable time in post-processing, at the expense of final image resolution. We present an analysis of the trade-offs in workflow efficiency and final image quality, and demonstrate that a few people with minimal equipment can efficiently digitize a teaching file library.

Current Clinical Practice Patterns of Self-Identified Interventional Radiologists

OBJECTIVE. The purpose of this study was to assess patterns of procedural, clinical evaluation and management (E/M), and diagnostic imaging services rendered by self-identified interventional radiologists (IRs) across the United States. MATERIALS AND METHODS. Recent Medicare Physician and Other Supplier Public Use and Physician Compare national downloadable files were linked. IRs were defined as physicians self-identifying interventional radiology as their primary specialty on Medicare claims or as a specialty during Medicare enrollment. The primary outcome measure was percentage of work (in work relative value units [WRVU]) attributed to interventional services (both procedural and E/M) per IR. Secondary outcome measures included sociodemographic factors per interventional WRVU quartile and percentage of E/M service units per IR. Statistical analysis included chi-square and t tests and logistic regression. RESULTS. Overall, 3132 physicians nationally self-identified to Medicare as IRs. The distribution o...

Automated Radiology Report Summarization Using an Open-Source Natural Language Processing Pipeline

Diagnostic radiologists are expected to review and assimilate findings from prior studies when constructing their overall assessment of the current study. Radiology information systems facilitate this process by presenting the radiologist with a subset of prior studies that are more likely to be relevant to the current study, usually by comparing anatomic coverage of both the current and prior studies. It is incumbent on the radiologist to review the full text report and/or images from those prior studies, a process that is time-consuming and confers substantial risk of overlooking a relevant prior study or finding. This risk is compounded when patients have dozens or even hundreds of prior imaging studies. Our goal is to assess the feasibility of natural language processing techniques to automatically extract asserted and negated disease entities from free-text radiology reports as a step towards automated report summarization. We compared automatically extracted disease mentions to a gold-standard set of manual annotations for 50 radiology reports from CT abdomen and pelvis examinations. The automated report summarization pipeline found perfect or overlapping partial matches for 86% of the manually annotated disease mentions (sensitivity 0.86, precision 0.66, accuracy 0.59, F1 score 0.74). The performance of the automated pipeline was good, and the overall accuracy was similar to the interobserver agreement between the two manual annotators.

De novo hepatocellular carcinoma occurrence in hepatitis C cirrhotics treated with direct-acting antiviral agents

Aim: Recent studies raise concerns for higher incidence of hepatocellular carcinoma (HCC) after direct-acting antiviral therapy for hepatitis C virus (HCV). Methods: In this study, using analysis of liver imaging pre- and post-DAA treatment, we queried new occurrence or ‘de novo’ of HCC in patients with HCV-cirrhosis treated with DAAs. Of 150 patients who met study criteria, 7 (4.7%; 95% CI: 2.1–9.5%) patients developed de novo HCC which did not differ from historical rates of 3% (p = 0.22). Results: Notably, patients with decompensated cirrhosis had significantly higher rate of de novo HCC (9.3%; 95% CI: 3.12–22.2%; p = 0.04). Conclusion: Our data support the need for continued surveillance for HCC in HCV cirrhotics even after successful therapy.

Incidentally detected bilateral adrenal nodules in patients without cancer: Is further workup necessary?

OBJECTIVE The purpose of this study was to determine the rate of malignancy in incidentally detected bilateral adrenal masses in patients with no known history of cancer. MATERIALS AND METHODS A retrospective search of CT reports of patients with incidentally detected bilateral adrenal nodules was performed from January 1, 2002, to January 1, 2014. Patients were excluded if they had a known cancer or suspected functioning adrenal tumor; 161 patients were included. Nodules were characterized as benign or malignant on the basis of imaging features at the index CT examination, imaging features at subsequent adrenal protocol CT or MRI, imaging stability for a minimum of 1 year, or clinical follow-up of a minimum of 2 years. RESULTS Mean nodule size was 1.8 cm (range, 0.7-4.9 cm). There were no cases of primary or secondary adrenal malignancy (95% CI, 0.00-0.023). The nodules diagnosed on index CT scans were 73 adrenal adenomas and two myelolipomas. Seventy-four nodules were subsequently characterized as adrenal adenomas on the basis of imaging findings. Of the 113 indeterminate nodules that had imaging follow-up, 111 were stable at the latest follow-up examination. One nodule grew 26% over 8.1 years, and the other grew 59% over 12.4 years. Clinical follow-up of patients with 60 indeterminate nodules revealed no evidence of adrenal malignancy. CONCLUSION No case of malignancy was found in 322 incidentally detected bilateral adrenal nodules at CT of patients without known cancer. Imaging follow-up of such lesions may be unnecessary.

Prevalence of Solid Tumors in Incidentally Detected Homogeneous Renal Masses Measuring > 20 HU on Portal Venous Phase CT

OBJECTIVE The purpose of this study was to determine the prevalence of solid tumors in incidental homogeneous renal masses with attenuation greater than 20 HU on portal venous phase CT images. MATERIALS AND METHODS In this retrospective study, the records of patients with incidental indeterminate (> 20 HU) homogeneous renal masses on portal venous phase CT scans from September 11, 2007, through March 18, 2017, were identified. Adult patients were included if they had undergone follow-up ultrasound, contrast-enhanced MRI, multiphase contrast-enhanced CT, or pathologic analysis alone to confirm the solid or cystic nature of the lesion. A single ROI was placed in the center of the mass, and lesions were characterized as ≥ 50% exophytic, < 50% exophytic, or entirely surrounded by renal parenchyma. RESULTS There were 322 masses in 267 patients. The mean lesion size was 16.6 (SD, 9.8) mm (range, 9-45 mm). Lesions were ≥ 50% exophytic in 92 cases, < 50% exophytic in 111 cases, and completely surrounded by renal parenchyma in 119 cases. All nonsolid lesions were characterized as benign cysts. The numbers of solid lesions per total number of lesions in each attenuation group were: 20-30 HU (0/140), 30-40 HU (0/67), 40-50 HU (1/38), 50-60 HU (3/24), 60-70 HU (5/17), 70-80 HU (5/17), and > 80 HU (8/19). All 207 lesions in the 20- to 40-HU range were benign cysts with no solid lesions (0%; 95% CI, 0.0-1.4%). CONCLUSION Small homogeneous renal masses measuring 20-40 HU on portal venous phase CT images are highly likely to be benign cysts.

PD28-10 MULTIPARAMETRIC-MRI PRIOR TO REPEAT BIOPSY FOR ACTIVE SURVEILLANCE CAN IDENTIFY MEN WITH HIGH-RISK PROSTATE CANCER

categories were reported in higher frequency in the patient-reported group compared to the physician-documented group: cardiovascular (25% vs. 20%), vascular-related (8.5% vs. 4.4%), neurologic (7.5% vs. 1.7%), gastrointestinal (30% vs. 25%), musculoskeletal comorbidities (30% vs. 21%), as well as other cancers (30% vs. 12%). Genitourinary comorbidities, including problems with urination and erectile dysfunction, were higher in the physician group (68% vs. 53%). CONCLUSIONS: Patients completing a medical history, at their own pace and in the comfort of their own home, provide relatively accurate and complete information, even before physician review. Electronic capturing of patient-reported comorbidities thus allows for an efficient method of obtaining a patient’s medical history and likely a more complete medical record.