RESUMEN
Purpose To determine an optimal embargo period preceding release of radiologic test results to an online patient portal. Materials and Methods This prospective discrete choice conjoint survey with modified orthogonal design was administered to patients by trained interviewers at four outpatient sites and two institutions from December 2016 to February 2018. Three preferences for receiving imaging results associated with a possible or known cancer diagnosis were evaluated: delay in receipt of results (1, 3, or 14 days), method of receipt (online portal, physician's office, or phone), and condition of receipt (before, at the same time as, or after health care provider). Preferences (hereafter, referred to as utilities) were derived from parameter estimates (ß) of multinomial regression stratified according to study participant and choice set. Results Among 464 screened participants, the response and completion rates were 90.5% (420 of 464) and 99.5% (418 of 420), respectively. Participants preferred faster receipt of results (P < .001) from their physician (P < .001) over the telephone (P < .001). Each day of delay decreased preference by 13 percentage points. Participants preferred immediate receipt of results through an online portal (utility, -.57) if made to wait more than 6 days to get results in the office and more than 11 days to get results by telephone. Compared with receiving results in their physician's office on day 7 (utility, -.60), participants preferred immediate release through the online portal without physician involvement if followed by a telephone call within 6 days (utility, -0.49) or an office visit within 2 days (utility, -.53). Older participants preferred physician-directed communication (P < .001). Conclusion The optimal embargo period preceding release of results through an online portal depends on the timing of traditional telephone- and office-based styles of communication. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Arenson et al in this issue.
Asunto(s)
Diagnóstico por Imagen , Registros Electrónicos de Salud , Neoplasias/diagnóstico por imagen , Acceso de los Pacientes a los Registros , Portales del Paciente , Adolescente , Adulto , Anciano , Niño , Femenino , Humanos , Masculino , Persona de Mediana Edad , Acceso de los Pacientes a los Registros/psicología , Acceso de los Pacientes a los Registros/estadística & datos numéricos , Prioridad del Paciente/psicología , Prioridad del Paciente/estadística & datos numéricos , Encuestas y Cuestionarios , Adulto JovenRESUMEN
Pulmonary embolism (PE) is a potentially lethal condition, and the diagnosis of PE can be difficult. The purpose of this study is to evaluate the incidence of PE on chest computed tomography angiography (CTA) studies ordered in the inpatient, outpatient, and emergency department (ED) settings and further segregated based on the adult and pediatric populations, and by the ordering clinician (attending physicians, resident physicians, or physician extenders). A retrospective review of chest CTA examinations performed between July 1,2009 and June 30, 2010 was performed. Of 5848 adult CTA studies, PE was diagnosed in 594 (10.1 %). Of these positive studies, 315 (53 %) were inpatient, 234 (39.4 %) were ED patients, and 45 (7.6 %) were outpatient. Four hundred sixty-four of 4445 (10.4 %) CTA examinations ordered by attending physicians were positive for PE. Seventy-four of the 801 (9.2 %) CTA examinations ordered by resident physicians were positive for PE. Fifty-six of the 608 CTA examinations ordered by physician extenders were positive for PE. Thirty-three pediatric CTA studies for PE met criteria and none of them indicated PE. There is no significant difference in the incidence of PE in chest CTA based on setting or ordering clinician.
Asunto(s)
Angiografía por Tomografía Computarizada , Embolia Pulmonar/diagnóstico por imagen , Adolescente , Adulto , Anciano , Anciano de 80 o más Años , Servicio de Urgencia en Hospital/estadística & datos numéricos , Femenino , Humanos , Masculino , Persona de Mediana Edad , Derivación y Consulta/estadística & datos numéricos , Estudios RetrospectivosRESUMEN
RATIONALE AND OBJECTIVES: Finding comparison to relevant prior studies is a requisite component of the radiology workflow. The purpose of this study was to evaluate the impact of a deep learning tool simplifying this time-consuming task by automatically identifying and displaying the finding in relevant prior studies. MATERIALS AND METHODS: The algorithm pipeline used in this retrospective study, TimeLens (TL), is based on natural language processing and descriptor-based image-matching algorithms. The dataset used for testing comprised 3872 series of 246 radiology examinations from 75 patients (189 CTs, 95 MRIs). To ensure a comprehensive testing, five finding types frequently encountered in radiology practice were included: aortic aneurysm, intracranial aneurysm, kidney lesion, meningioma, and pulmonary nodule. After a standardized training session, nine radiologists from three university hospitals performed two reading sessions on a cloud-based evaluation platform resembling a standard RIS/PACS. The task was to measure the diameter of the finding-of-interest on two or more exams (a most recent and at least one prior exam): first without use of TL, and a second session at an interval of at least 21 days with the use of TL. All user actions were logged for each round, including time needed to measure the finding at all timepoints, number of mouse clicks, and mouse distance traveled. The effect of TL was evaluated in total, per finding type, per reader, per experience (resident vs. board-certified radiologist), and per modality. Mouse movement patterns were analyzed with heatmaps. To assess the effect of habituation to the cases, a third round of readings was performed without TL. RESULTS: Across scenarios, TL reduced the average time needed to assess a finding at all timepoints by 40.1% (107 vs. 65 seconds; p < 0.001). Largest accelerations were demonstrated for assessment of pulmonary nodules (-47.0%; p < 0.001). Less mouse clicks (-17.2%) were needed for finding evaluation with TL, and mouse distance traveled was reduced by 38.0%. Time needed to assess the findings increased from round 2 to round 3 (+27.6%; p < 0.001). Readers were able to measure a given finding in 94.4% of cases on the series initially proposed by TL as most relevant series for comparison. The heatmaps showed consistently simplified mouse movement patterns with TL. CONCLUSION: A deep learning tool significantly reduced both the amount of user interactions with the radiology image viewer and the time needed to assess findings of interest on cross-sectional imaging with relevant prior exams.
Asunto(s)
Aprendizaje Profundo , Humanos , Estudios Retrospectivos , Radiólogos , Imagen por Resonancia Magnética/métodos , Tomografía Computarizada por Rayos X/métodosRESUMEN
In modern clinical decision-support algorithms, heterogeneity in image characteristics due to variations in imaging systems and protocols hinders the development of reproducible quantitative measures including for feature extraction pipelines. With the help of a reader study, we investigate the ability to provide consistent ground-truth targets by using patient-specific 3D-printed lung phantoms. PixelPrint was developed for 3D-printing lifelike computed tomography (CT) lung phantoms by directly translating clinical images into printer instructions that control density on a voxel-by-voxel basis. Data sets of three COVID-19 patients served as input for 3D-printing lung phantoms. Five radiologists rated patient and phantom images for imaging characteristics and diagnostic confidence in a blinded reader study. Effect sizes of evaluating phantom as opposed to patient images were assessed using linear mixed models. Finally, PixelPrint's production reproducibility was evaluated. Images of patients and phantoms had little variation in the estimated mean (0.03-0.29, using a 1-5 scale). When comparing phantom images to patient images, effect size analysis revealed that the difference was within one-third of the inter- and intrareader variabilities. High correspondence between the four phantoms created using the same patient images was demonstrated by PixelPrint's production repeatability tests, with greater similarity scores between high-dose acquisitions of the phantoms than between clinical-dose acquisitions of a single phantom. We demonstrated PixelPrint's ability to produce lifelike CT lung phantoms reliably. These phantoms have the potential to provide ground-truth targets for validating the generalizability of inference-based decision-support algorithms between different health centers and imaging protocols and for optimizing examination protocols with realistic patient-based phantoms. Classification: CT lung phantoms, reader study.
RESUMEN
BACKGROUND: Recent advances in performance of natural language processing (NLP) techniques have spurred wider use and more sophisticated applications of NLP in radiology. This study systematically reviews the trends and applications of NLP in radiology within the last five years. METHODS: A search of three databases of peer-reviewed journal articles and conference papers from January 1, 2016 to April 21, 2021 resulted in a total of 228 publications included in the review. Manuscripts were analyzed by several factors, including clinical application, study setting, NLP technique, and performance. RESULTS: Of the 228 included publications, there was an overall increase in number of studies published with an increase in use of machine learning models. NLP models showed high performance: >50% of publications reported F1 > 0.91. There was variable sample size across the studies with a median of 3708 data points, most commonly radiology reports. 145 studies utilized data from a single academic center. Applications were classified as clinical (n = 87), technical (n = 66), quality improvement (n = 61), research (n = 9), and education (n = 5). DISCUSSION: There has been a continued increase in number of studies involving NLP in radiology. Newer NLP techniques, including word embedding, deep learning, and transformers, are being applied and show improved performance. There has been growth in the interpretative and non-interpretative use of NLP techniques in radiology and has great capacity to improve patient care and delivery. Although the performance and breadth of NLP applications is impressive, there is an overall lack of high-level evidence for actual clinical application of published tools. CONCLUSION: NLP applications in radiology has been increasing studied and more accurate in the last 5 years. More direct clinical application and portability of the NLP pipelines is need to reach the technology's full potential.
Asunto(s)
Procesamiento de Lenguaje Natural , Radiología , Manejo de Datos , Humanos , Aprendizaje Automático , RadiografíaRESUMEN
STUDY DESIGN: Retrospective. OBJECTIVE: To compare the rate of positive pathology on thoracic MRI ordered by surgical spine specialists to those ordered by nonsurgical spine specialists. METHODS: Outpatient thoracic MRIs from January-March 2019 were evaluated from a single academic health care system. Studies without a known ordering provider, imaging report, or patients with known presence of malignancy, multiple sclerosis, recent trauma, or surgery were excluded (n = 320). Imaging studies were categorized by type of provider placing the order (resident, attending, or advanced practice practitioner) and department. MRIs were deemed positive if they showed relevant pathology that correlated with indication for exam as determined by a radiologist. One-sided chi-squared analysis was performed to determine statistical significance. RESULTS: Overall, our data demonstrated 17.2% of studies with positive pathology. Compared to nonspecialty clinicians, subspecialists showed 35/184 (19.0%) positivity rate versus the non-specialist with 20/136 (14.7%) positivity rate (P = .156). Posthoc analysis demonstrated that surgical specialists who order thoracic MRIs yield significantly higher positivity rates at 19/79 (24.0%) compared to nonsurgical specialists at 36/241 (14.9%) (P < .05). Overall, neurosurgery demonstrated the highest rate of positive thoracic MRIs at 14/40 (35.0%). Comparison between the rate of positivity between physicians and advanced practitioners was insignificant (P > .05). CONCLUSIONS: Clinical diagnosis of symptomatic thoracic spine degenerative disease requires an expert physical exam combined with careful attention to radiology findings. Although the percent of relevant pathology on thoracic MRI is low, our data suggests evaluation by a surgical specialist should precede ordering a thoracic spine MRI.