Paras Lakhani

Deep Learning at Chest Radiography: Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural Networks

Purpose To evaluate the efficacy of deep convolutional neural networks (DCNNs) for detecting tuberculosis (TB) on chest radiographs. Materials and Methods Four deidentified HIPAA-compliant datasets were used in this study that were exempted from review by the institutional review board, which consisted of 1007 posteroanterior chest radiographs. The datasets were split into training (68.0%), validation (17.1%), and test (14.9%). Two different DCNNs, AlexNet and GoogLeNet, were used to classify the images as having manifestations of pulmonary TB or as healthy. Both untrained and pretrained networks on ImageNet were used, and augmentation with multiple preprocessing techniques. Ensembles were performed on the best-performing algorithms. For cases where the classifiers were in disagreement, an independent board-certified cardiothoracic radiologist blindly interpreted the images to evaluate a potential radiologist-augmented workflow. Receiver operating characteristic curves and areas under the curve (AUCs) were used to assess model performance by using the DeLong method for statistical comparison of receiver operating characteristic curves. Results The best-performing classifier had an AUC of 0.99, which was an ensemble of the AlexNet and GoogLeNet DCNNs. The AUCs of the pretrained models were greater than that of the untrained models (P < .001). Augmenting the dataset further increased accuracy (P values for AlexNet and GoogLeNet were .03 and .02, respectively). The DCNNs had disagreement in 13 of the 150 test cases, which were blindly reviewed by a cardiothoracic radiologist, who correctly interpreted all 13 cases (100%). This radiologist-augmented approach resulted in a sensitivity of 97.3% and specificity 100%. Conclusion Deep learning with DCNNs can accurately classify TB at chest radiography with an AUC of 0.99. A radiologist-augmented approach for cases where there was disagreement among the classifiers further improved accuracy.

Is Android or iPhone the Platform for Innovation in Imaging Informatics

It is clear that ubiquitous mobile computing platforms will be a disruptive technology in the delivery of healthcare in the near future. While radiologists are fairly sedentary, their customers, the referring physicians, and the patients are not. The need for closer collaboration and interaction with referring physicians is seen as a key to maintaining relationships and integrating tightly with the patient management team. While today, patients have to settle for their images on a CD, in short time, they will be taking them home on their cell phone. As PACS vendors are moving ever outward in the enterprise, they are already actively developing clients on mobile platforms. Two major contenders are the Apple’s iPhone and the Android platform developed by Google. These two designs represent two entirely different architectures and business models.

Automated Detection of Critical Results in Radiology Reports

The goal of this study was to develop and validate text-mining algorithms to automatically identify radiology reports containing critical results including tension or increasing/new large pneumothorax, acute pulmonary embolism, acute cholecystitis, acute appendicitis, ectopic pregnancy, scrotal torsion, unexplained free intraperitoneal air, new or increasing intracranial hemorrhage, and malpositioned tubes and lines. The algorithms were developed using rule-based approaches and designed to search for common words and phrases in radiology reports that indicate critical results. Certain text-mining features were utilized such as wildcards, stemming, negation detection, proximity matching, and expanded searches with applicable synonyms. To further improve accuracy, the algorithms utilized modality and exam-specific queries, searched under the “Impression” field of the radiology report, and excluded reports with a low level of diagnostic certainty. Algorithm accuracy was determined using precision, recall, and F-measure using human review as the reference standard. The overall accuracy (F-measure) of the algorithms ranged from 81% to 100%, with a mean precision and recall of 96% and 91%, respectively. These algorithms can be applied to radiology report databases for quality assurance and accreditation, integrated with existing dashboards for display and monitoring, and ported to other institutions for their own use.

Radiology Reporting and Communications: A Look Forward

The content and prose method of radiology reporting has remained essentially unchanged for more than 100 years. By leveraging current technologies, the radiology report has the potential to be a multifunctional document providing information in a number of areas including business analytics, quality assurance and safety, regulatory reporting, research and billing. Maturation and adoption of speech recognition, the development of radiology controlled terminologies and standardized reporting templates now allow for the introduction of structured reporting into the clinical setting.

Detection of Endoleaks after Endovascular Aneurysm Repair with Use of Technetium-99m Sulfur Colloid and 99m Tc-labeled Red Blood Cell Scans

PURPOSE This study was performed to determine whether endoleaks could be detected after endovascular aneurysm repair (EVAR) with use of technetium-99m sulfur colloid and (99m)Tc-labeled red blood cell (RBC) nuclear medicine scans. MATERIALS AND METHODS There were 13 patients enrolled in this study: nine with endoleaks seen on computed tomographic (CT) angiography and four with no endoleak on CT angiography. All patients underwent regularly scheduled surveillance CT angiography examination after EVAR to evaluate for endoleak. Endoleak detection was then attempted in each patient with two nuclear medicine scans: a (99m)Tc sulfur colloid scan and a (99m)Tc-labeled RBC scan. Flow images (5 seconds per frame) were obtained for 1 minute after intravenous administration of 555 MBq (15 mCi) (99m)Tc sulfur colloid. Sequential dynamic images were then obtained every minute for 30 minutes. Next, a (99m)Tc-labeled RBC study was performed after the intravenous administration of 370-1,073 MBq (10-29 mCi) in vitro labeled (99m)Tc RBCs. Flow images were obtained, followed by sequential dynamic images obtained every minute for 30 minutes. Single photon emission CT images of the abdomen were then acquired. The nuclear medicine scans were evaluated for the presence or absence of endoleak independent of the CT angiography findings. RESULTS Of the nine patients with endoleaks on CT angiography, seven (78%) had them detected by nuclear medicine examinations. Two of the nine endoleaks seen on CT angiography (22%) were not seen on either scintigraphic examination. All patients with no endoleak on CT angiography had their nuclear medicine scans correctly interpreted as showing no endoleak present (n = 4; 100%). No complications occurred as a result of the nuclear medicine scans. CONCLUSIONS Endoleaks can be detected with (99m)Tc sulfur colloid and (99m)Tc-labeled RBC nuclear medicine scans. This initial work suggests that the sensitivities of these scintigraphic scanning methods for endoleak detection are lower than that of CT angiography.

Automated Detection of Radiology Reports that Document Non-routine Communication of Critical or Significant Results

The purpose of this investigation is to develop an automated method to accurately detect radiology reports that indicate non-routine communication of critical or significant results. Such a classification system would be valuable for performance monitoring and accreditation. Using a database of 2.3 million free-text radiology reports, a rule-based query algorithm was developed after analyzing hundreds of radiology reports that indicated communication of critical or significant results to a healthcare provider. This algorithm consisted of words and phrases used by radiologists to indicate such communications combined with specific handcrafted rules. This algorithm was iteratively refined and retested on hundreds of reports until the precision and recall did not significantly change between iterations. The algorithm was then validated on the entire database of 2.3 million reports, excluding those reports used during the testing and refinement process. Human review was used as the reference standard. The accuracy of this algorithm was determined using precision, recall, and F measure. Confidence intervals were calculated using the adjusted Wald method. The developed algorithm for detecting critical result communication has a precision of 97.0% (95% CI, 93.5–98.8%), recall 98.2% (95% CI, 93.4–100%), and F measure of 97.6% (ß = 1). Our query algorithm is accurate for identifying radiology reports that contain non-routine communication of critical or significant results. This algorithm can be applied to a radiology reports database for quality control purposes and help satisfy accreditation requirements.

Computer Input Devices: Neutral Party or Source of Significant Error in Manual Lesion Segmentation?

Lesion segmentation involves outlining the contour of an abnormality on an image to distinguish boundaries between normal and abnormal tissue and is essential to track malignant and benign disease in medical imaging for clinical, research, and treatment purposes. A laser optical mouse and a graphics tablet were used by radiologists to segment 12 simulated reference lesions per subject in two groups (one group comprised three lesion morphologies in two sizes, one for each input device for each device two sets of six, composed of three morphologies in two sizes each). Time for segmentation was recorded. Subjects completed an opinion survey following segmentation. Error in contour segmentation was calculated using root mean square error. Error in area of segmentation was calculated compared to the reference lesion. 11 radiologists segmented a total of 132 simulated lesions. Overall error in contour segmentation was less with the graphics tablet than with the mouse (P < 0.0001). Error in area of segmentation was not significantly different between the tablet and the mouse (P = 0.62). Time for segmentation was less with the tablet than the mouse (P = 0.011). All subjects preferred the graphics tablet for future segmentation (P = 0.011) and felt subjectively that the tablet was faster, easier, and more accurate (P = 0.0005). For purposes in which accuracy in contour of lesion segmentation is of the greater importance, the graphics tablet is superior to the mouse in accuracy with a small speed benefit. For purposes in which accuracy of area of lesion segmentation is of greater importance, the graphics tablet and mouse are equally accurate.

Documentation of Nonroutine Communications of Critical or Significant Radiology Results: A Multiyear Experience at a Tertiary Hospital

PURPOSE The aim of this study was to determine the frequency of radiology reports that contain nonroutine communications of results and categorize the urgency of such communications. METHODS A rule-based text-query algorithm was applied to a database of 2.3 million radiology reports, which has an accuracy of 98% for classifying reports containing documentation of communications. The frequency of such communications by year, modality, and study type was then determined. Finally, 200 random reports selected by the algorithm were analyzed, and reports containing critical results were categorized according to ascending levels of urgency. RESULTS Critical or noncritical results to health care providers were present in 5.09% of radiology reports (116,184 of 2,282,923). For common modalities, documentation of communications were most frequent in CT (14.34% [57,537 of 402,060]), followed by ultrasound (9.55% [17,814 of 186,626]), MRI (5.50% [13,697 of 248,833]), and chest radiography (1.57% [19,840 of 1,262,925]). From 1997 to 2005, there was an increase in reports containing such communications (3.04% in 1997, 6.82% in 2005). More reports contained nonroutine communications in single-view chest radiography (1.29% [5,533 of 428,377]) than frontal/lateral chest radiography (0.80% [1,815 of 226,837]), diagnostic mammography (9.42% [3,662 of 38,877]) than screening mammography (0.47% [289 of 61,114]), and head CT (26.21% [20,963 of 79,985]) than abdominal CT (15.05% [19,871 of 132,034]) or chest CT (5.33% [3,017 of 56,613]). All of these results were statistically significant (P < .00001). Of 200 random radiology reports indicating nonroutine communications, 155 (78%) had critical and 45 (22%) had noncritical results. Regarding level of urgency, 94 of 155 reports (60.6%) with critical results were categorized as high urgency, 31 (20.0%) as low urgency, 26 (16.8%) as medium urgency, and 4 (2.6%) as discrepant. CONCLUSIONS From 1997 to 2005, there was a significant increase in documentation of nonroutine communications, which may be due to increasing compliance with ACR guidelines. Most reports with nonroutine communications contain critical findings.

Development and Validation of Queries Using Structured Query Language (SQL) to Determine the Utilization of Comparison Imaging in Radiology Reports Stored on PACS

The purpose of this research was to develop queries that quantify the utilization of comparison imaging in free-text radiology reports. The queries searched for common phrases that indicate whether comparison imaging was utilized, not available, or not mentioned. The queries were iteratively refined and tested on random samples of 100 reports with human review as a reference standard until the precision and recall of the queries did not improve significantly between iterations. Then, query accuracy was assessed on a new random sample of 200 reports. Overall accuracy of the queries was 95.6%. The queries were then applied to a database of 1.8 million reports. Comparisons were made to prior images in 38.69% of the reports (693,955/1,793,754), were unavailable in 18.79% (337,028/1,793,754), and were not mentioned in 42.52% (762,771/1,793,754). The results show that queries of text reports can achieve greater than 95% accuracy in determining the utilization of prior images.

Automated Extraction of Critical Test Values and Communications from Unstructured Radiology Reports: An Analysis of 9.3 Million Reports from 1990 to 2011

PURPOSE To determine the frequency of critical radiology results in 9.3 million radiology reports from our health system, to identify those containing documentation of communication by using automated text-classification algorithms, and to assess the impact of a policy requiring documentation of critical results communication. MATERIALS AND METHODS This HIPAA-compliant retrospective study received institutional review board approval. Text-mining algorithms that were previously validated to have mean accuracies of more than 90% for identifying certain critical results and documentation of communications were applied to a database of 9.3 million radiology reports. The frequency of critical results and documentation of communication were then determined from 1990 to 2011. RESULTS There was an increase in documentation of communication for all critical results from 1990 to 2011. In 1990, 19.0% of reports with critical values had evidence of documentation of communication compared with 72.4% of reports in 2010. The linear trend for increasing documentation of communications began in 1997 and continued until 2011 (P < .001). From 1990 to 2011, documentation of communication was highest in acute scrotal torsion (70.6%) and ectopic pregnancy (65.4%) and lowest in unexplained free-intraperitoneal air (29.5%) and malpositioned tubes (30.4%). In 2010-2011, radiologists were least likely to document communication of results for malpositioned endotracheal and enteric tubes (2010, 58.56%; 2011, 57.50%) and unexplained free-intraperitoneal air (2010, 59.57%; 2011, 75.51%). They were most likely to document communication of results for ectopic pregnancy (2010, 94.12%; 2011, 93.48%) and acute appendicitis (2010, 86.87%; 2011, 84.31%). CONCLUSION There was an increase in documentation of communication of critical results, which demonstrated a rising linear trend that began in 1997 and continued until 2011. The increasing trend began well before policy implementation, indicating that other factors such as heightened awareness among radiologists likely had a role.