Automated identification and assignment of colonoscopy surveillance recommendations for individuals with colorectal polyps.

BACKGROUND AND AIMS: Determining surveillance intervals for patients with colorectal polyps is critical but time-consuming and challenging to do reliably. We present the development and assessment of a pipeline that leverages natural language processing techniques to automatically extract and analyze relevant polyp findings from free-text colonoscopy and pathology reports. Using this information, we categorized individual patients into 6 postcolonoscopy surveillance intervals defined by the U.S. Multi-Society Task Force on Colorectal Cancer. METHODS: Using a set of 546 randomly selected colonoscopy and pathology reports from 324 patients in a single health system, we used a combination of statistical classifiers and rule-based methods to extract polyp properties from each report type, associate properties with unique polyps, and classify a patient into 1 of 6 risk categories by integrating information from both report types. We then assessed the pipeline's performance by determining the positive predictive value (PPV), sensitivity, and F-score of the algorithm, compared with the determination of surveillance intervals by a gastroenterologist. RESULTS: The pipeline was developed using 346 reports (224 colonoscopy and 122 pathology) from 224 patients and evaluated on an independent test set of 200 reports (100 colonoscopy and 100 pathology) from 100 patients. We achieved an average PPV, sensitivity, and F-score of .92, .95, and .93, respectively, across targeted entities for colonoscopy. Pathology extraction achieved a PPV, sensitivity, and F-score of .95, .97, and .96. The system achieved an overall accuracy of 92% in assigning the recommended interval for surveillance colonoscopy. CONCLUSIONS: This study demonstrates the feasibility of using machine learning to automatically extract findings and classify patients to appropriate risk categories and corresponding surveillance intervals. Incorporating this system can facilitate proactive and timely follow-up after screening colonoscopy and enable real-time quality assessment of prevention programs and providers.

Gastroenterology visitation and reminders predict surveillance uptake for patients with adenomas with high-risk features.

Individuals diagnosed with colorectal adenomas with high-risk features during screening colonoscopy have increased risk for the development of subsequent adenomas and colorectal cancer. While US guidelines recommend surveillance colonoscopy at 3 years in this high-risk population, surveillance uptake is suboptimal. To inform future interventions to improve surveillance uptake, we sought to assess surveillance rates and identify facilitators of uptake in a large integrated health system. We utilized a cohort of patients with a diagnosis of ≥ 1 tubular adenoma (TA) with high-risk features (TA ≥ 1 cm, TA with villous features, TA with high-grade dysplasia, or ≥ 3 TA of any size) on colonoscopy between 2013 and 2016. Surveillance colonoscopy completion within 3.5 years of diagnosis of an adenoma with high-risk features was our primary outcome. We evaluated surveillance uptake over time and utilized logistic regression to detect factors associated with completion of surveillance colonoscopy. The final cohort was comprised of 405 patients. 172 (42.5%) patients successfully completed surveillance colonoscopy by 3.5 years. Use of a patient reminder (telephone, electronic message, or letter) for due surveillance (adjusted odds = 1.9; 95%CI = 1.2-2.8) and having ≥ 1 gastroenterology (GI) visit after diagnosis of an adenoma with high-risk features (adjusted odds = 2.6; 95%CI = 1.6-4.2) significantly predicted surveillance colonoscopy completion at 3.5 years. For patients diagnosed with adenomas with high-risk features, surveillance colonoscopy uptake is suboptimal and frequently occurs after the 3-year surveillance recommendation. Patient reminders and visitation with GI after index colonoscopy are associated with timely surveillance completion. Our findings highlight potential health system interventions to increase timely surveillance uptake for patients diagnosed with adenomas with high-risk features.

Influence of the Location and Zone of Tumor in Prostate Cancer Detection and Localization on 3-T Multiparametric MRI Based on PI-RADS Version 2.

OBJECTIVE. The objective of our study was to determine the performance of 3-T multiparametric MRI (mpMRI) for prostate cancer (PCa) detection and localization, stratified by anatomic zone and level, using Prostate Imaging Reporting and Data System version 2 (PI-RADSv2) and whole-mount histopathology (WMHP) as reference. MATERIALS AND METHODS. Multiparametric MRI examinations of 415 consecutive men were compared with thin-section WMHP results. A genitourinary radiologist and pathologist collectively determined concordance. Two radiologists assigned PI-RADSv2 scores and sector location to all detected foci by consensus. Tumor detection rates were calculated for clinical and pathologic (tumor location and zone) variables. Both rigid and adjusted sector-matching models were used to account for fixation-related issues. RESULTS. Of 863 PCa foci in 16,185 prostate sectors, the detection of overall and index PCa lesions in the midgland, base, and apex was 54.9% and 83.1%, 42.1% and 64.0% (p = 0.04, p = 0.02), and 41.9% and 71.4% (p = 0.001, p = 0.006), respectively. Tumor localization sensitivity was highest in the midgland compared with the base and apex using an adjusted match compared with a rigid match (index lesions, 71.3% vs 43.7%; all lesions, 70.8% vs 36.0%) and was greater in the peripheral zone (PZ) than in the transition zone. Three-Tesla mpMRI had similarly high specificity (range, 93.8-98.3%) for overall and index tumor localization when using both rigid and adjusted sector-matching approaches. CONCLUSION. For 3-T mpMRI, the highest sensitivity (83.1%) for detection of index PCa lesions was in the midgland, with 98.3% specificity. Multiparametric MRI performance for sectoral localization of PCa within the prostate was moderate and was best for index lesions in the PZ using an adjusted model.

Detection and Localization of Prostate Cancer at 3-T Multiparametric MRI Using PI-RADS Segmentation.

OBJECTIVE. The purpose of this study is to determine the overall and sector-based performance of 3-T multiparametric MRI for prostate cancer (PCa) detection and localization by using Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) scoring and segmentation compared with whole-mount histopathologic analysis. MATERIALS AND METHODS. Multiparametric 3-T MRI examinations of 415 consecutive men were compared with thin-section whole-mount histopathologic analysis. A genitourinary radiologist and pathologist collectively determined concordance. Two radiologists assigned PI-RADSv2 categories and sectoral location to all detected foci by consensus. Tumor detection rates were calculated for clinical and pathologic (Gleason score) variables. Both rigid and adjusted sector-matching models were used to account for fixation-related issues. RESULTS. The 415 patients had 863 PCa foci (52.7% had a Gleason score ≥ 7, 61.9% were ≥ 1 cm, and 90.4% (375/415) of index lesions were ≥ 1 cm) and 16,185 prostate sectors. Multiparametric MRI enabled greater detection of PCa lesions 1 cm or larger (all lesions vs index lesions, 61.6% vs 81.6%), lesions with Gleason score greater than or equal to 7 (all lesions vs index lesions, 71.4% vs 80.9%), and index lesions with both Gleason score greater than or equal to 7 and size 1 cm or larger (83.3%). Higher sensitivity was obtained for adjusted versus rigid tumor localization for all lesions (56.0% vs 28.5%), index lesions (55.4% vs 34.3%), lesions with Gleason score greater than or equal to 7 (55.7% vs 36.0%), and index lesions 1 cm or larger (56.1% vs 35.0%). Multiparametric 3-T MRI had similarly high specificity (96.0-97.9%) for overall and index tumor localization with adjusted and rigid sector-matching approaches. CONCLUSION. Using 3-T multiparametric MRI and PI-RADSv2, we achieved the highest sensitivity (83.3%) for the detection of lesions 1 cm or larger with Gleason score greater than or equal to 7. Sectoral localization of PCa within the prostate was moderate and was better with an adjusted model than with a rigid model.

Detection of Individual Prostate Cancer Foci via Multiparametric Magnetic Resonance Imaging.

BACKGROUND: Multiparametric magnetic resonance imaging (mpMRI) undoubtedly affects the diagnosis and treatment of localized prostate cancer (CaP). However, clinicians need a better understanding of its accuracy and limitations in detecting individual CaP foci to optimize management. OBJECTIVE: To determine the per-lesion detection rate for CaP foci by mpMRI and identify predictors of tumor detection. DESIGN, SETTING, AND PARTICIPANTS: We carried out a retrospective analysis of a prospectively managed database correlating lesion-specific results from mpMRI co-registered with whole-mount pathology (WMP) prostatectomy specimens from June 2010 to February 2018. Participants include 588 consecutive patients with biopsy-proven CaP undergoing 3-T mpMRI before radical prostatectomy at a single tertiary institution. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: We measured mpMRI sensitivity in detecting individual CaP and clinically significant (any Gleason score ≥7) CaP foci and predictors of tumor detection using multivariate analysis. RESULTS AND LIMITATIONS: The final analysis included 1213 pathologically confirmed tumor foci in 588 patients with primarily intermediate- (75%) or high-risk (12%) CaP. mpMRI detected 45% of all lesions (95% confidence interval [CI] 42-47%), including 65% of clinically significant lesions (95% CI 61-69%) and nearly 80% of high-grade tumors. Some 74% and 31% of missed solitary and multifocal tumors, respectively, were clinically significant. The majority of missed lesions were small (61.1% ≤1cm); 28.3% were between 1 and 2cm, and 10.4% were >2cm. mpMRI missed at least one clinically significant focus in 34% of patients overall, and in 45% of men with multifocal lesions. On multivariate analysis, smaller, low-grade, multifocal, nonindex tumors with lower prostate-specific antigen density were more likely to be missed. Limitations include selection bias in a prostatectomy cohort, lack of specificity data, an imperfect co-registration process, and uncertain clinical significance for undetected lesions. CONCLUSIONS: mpMRI detects less than half of all and less than two-thirds of clinically significant CaP foci. The moderate per-lesion sensitivity and significant proportion of men with undetected tumor foci demonstrate the current limitations of mpMRI. PATIENT SUMMARY: Magnetic resonance imaging of the prostate before surgical removal for prostate cancer finds less than half of all individual prostate cancer tumors. Large, solitary, aggressive tumors are more likely to be visualized on imaging.

Building a Natural Language Processing Tool to Identify Patients With High Clinical Suspicion for Kawasaki Disease from Emergency Department Notes.

OBJECTIVE: Delayed diagnosis of Kawasaki disease (KD) may lead to serious cardiac complications. We sought to create and test the performance of a natural language processing (NLP) tool, the KD-NLP, in the identification of emergency department (ED) patients for whom the diagnosis of KD should be considered. METHODS: We developed an NLP tool that recognizes the KD diagnostic criteria based on standard clinical terms and medical word usage using 22 pediatric ED notes augmented by Unified Medical Language System vocabulary. With high suspicion for KD defined as fever and three or more KD clinical signs, KD-NLP was applied to 253 ED notes from children ultimately diagnosed with either KD or another febrile illness. We evaluated KD-NLP performance against ED notes manually reviewed by clinicians and compared the results to a simple keyword search. RESULTS: KD-NLP identified high-suspicion patients with a sensitivity of 93.6% and specificity of 77.5% compared to notes manually reviewed by clinicians. The tool outperformed a simple keyword search (sensitivity = 41.0%; specificity = 76.3%). CONCLUSIONS: KD-NLP showed comparable performance to clinician manual chart review for identification of pediatric ED patients with a high suspicion for KD. This tool could be incorporated into the ED electronic health record system to alert providers to consider the diagnosis of KD. KD-NLP could serve as a model for decision support for other conditions in the ED.

Ensembles of NLP Tools for Data Element Extraction from Clinical Notes.

Natural Language Processing (NLP) is essential for concept extraction from narrative text in electronic health records (EHR). To extract numerous and diverse concepts, such as data elements (i.e., important concepts related to a certain medical condition), a plausible solution is to combine various NLP tools into an ensemble to improve extraction performance. However, it is unclear to what extent ensembles of popular NLP tools improve the extraction of numerous and diverse concepts. Therefore, we built an NLP ensemble pipeline to synergize the strength of popular NLP tools using seven ensemble methods, and to quantify the improvement in performance achieved by ensembles in the extraction of data elements for three very different cohorts. Evaluation results show that the pipeline can improve the performance of NLP tools, but there is high variability depending on the cohort.

Journal club: Renal masses detected at abdominal CT: radiologists' adherence to guidelines regarding management recommendations and communication of critical results.

OBJECTIVE: The purpose of this study was to assess radiologists' adherence to published guidelines for managing renal masses detected at abdominal CT at one institution and to a critical results communication policy. MATERIALS AND METHODS: A validated natural language processing tool supplemented by manual review was used to randomly assemble a cohort of 97 radiology reports from all abdominal CT reports (n = 11,952) generated from July 2010 to June 2011. Critical renal mass findings warranted consideration for surgery, intervention, or imaging follow-up and required direct, separate, and timely communication to the referrer in addition to the radiology report. Primary outcomes were adherence to guidelines and institutional policy for communicating critical results. Sample size allowed a 95% CI ± 5% for primary outcome. Pearson chi-square test was performed to assess whether radiology subspecialization was predictive of the primary outcome. RESULTS: Of all abdominal CT reports, 35.6% contained at least one renal mass finding (4.3% critical). Guideline adherence was lower for patients with critical than for those with noncritical findings (48/57 [84.2%] vs 40/40 [100%]; p = 0.01). Adherence to critical result communication policy was 73.7% (42/57). For critical findings, abdominal radiologists had higher guideline adherence (40/43 [93.0%] vs 8/14 [57.1%]; p = 0.001) and critical result communication policy adherence (36/43 [83.7%] vs 6/14 [42.9%]; p = 0.002) than nonabdominal radiologists. CONCLUSION: In reporting renal masses detected at abdominal CT, radiologists largely adhered to management guidelines but did not adhere to the critical results communication policy in one of four reports. Subspecialization improved adherence to both management guidelines and the institution's critical result communication policy.

Simple cyst-appearing renal masses at unenhanced CT: can they be presumed to be benign?

PURPOSE: To determine renal cancer incidence in simple cyst-appearing renal masses detected at unenhanced computed tomography (CT). MATERIALS AND METHODS: Institutional review board approval and an informed consent waiver for this retrospective HIPAA-compliant study were obtained. Patients who had renal masses with homogeneous water attenuation, hairline-thin smooth walls, and no calcifications or septations were identified by applying a validated natural language processing algorithm to radiology reports for 15 695 unique patients who underwent unenhanced abdominal CT at our institution between 2000 and 2005. Reports that included renal masses were selected, then categorized through manual report review as pertaining to simple cyst-appearing renal masses, nonsimple or solid renal masses, or no renal masses. Medical records were reviewed for subsequent renal cancer diagnoses. Patients without renal cancer were evaluated for a minimum of 5 years (mean, 8 years; range, 5-12 years). The Cox proportional hazards regression model was used to compare renal cancer incidence for patients who had simple cyst-appearing renal masses with those who had nonsimple cystic or solid renal masses and those who had no renal masses. RESULTS: Simple cyst-appearing renal masses were identified in 2669 patients (17%), no renal masses in 11844 (75%), and nonsimple cystic or solid renal masses in 1182 (8%). Of 1159 patients with simple cyst-appearing renal masses and a minimum of 5 years of follow-up, six (0.52%) subsequently developed renal cancers, all of which were separate from the simple cyst-appearing renal mass, rather than within it. Of 446 patients with nonsimple or solid renal masses and sufficient follow-up, 50 (11%) developed renal cancer. There was no difference in renal cancer incidence in patients with simple cyst-appearing renal masses versus those without renal masses (P = .54). The incidence of renal cancer was significantly lower in patients with simple cyst-appearing renal masses than that in nonsimple cystic or solid renal masses (P < .0001). CONCLUSION: Simple cyst-appearing renal masses are unlikely to be malignant. These data support foregoing further imaging evaluation of these common masses.

Utilization effect of integrating a chest radiography room into a thoracic surgery ward.

PURPOSE: Bedside chest radiography (CXR) represents a substantial fraction of the volume of medical imaging for inpatient health care facilities. However, its image quality is limited compared with posterior-anterior/lateral (PA/LAT) acquisitions taken in radiographic rooms. The aim of this study was to evaluate the utilization of bedside CXR and other chest imaging modalities before and after placing a radiography room within a thoracic surgical inpatient ward. METHODS: All patient admissions (n = 3,852) to the thoracic surgical units between April 1, 2007, and December 31, 2010, were retrospectively identified. All chest imaging tests performed for these patients, including CT scans, MRI, ultrasound, and bedside and PA/LAT radiography, were counted. The primary outcome measure was chest imaging utilization, defined as the number of chest examinations per admission, before and after the establishment of the digital radiography room on January 10, 2010. Statistical analysis was performed using an independent-samples t test to evaluate changes in chest imaging utilization. RESULTS: A 2.61-fold increase in the number of PA/LAT CXR studies per admission (P < .01) and a 1.96-fold decrease in the number of bedside CXR studies per admission (P < .01) were observed after radiography room implementation. The number of chest CT, MRI, and ultrasound studies per admission did not change significantly. CONCLUSIONS: Establishing a radiography room physically within thoracic surgery units or in close proximity can significantly shift CXR utilization from bedside to PA/LAT acquisitions, which may enable opportunities for improvement in efficiency, quality, and safety in patient care.