Current Problems in Diagnostic Radiology最新文献

Introduction

The construction and results of a multiple-reader multiple-case prostate MRI study are described and reported to illustrate recommendations for how to standardize artificial intelligence (AI) prostate studies per the review constituting Part I¹.

Methods

Our previously reported approach was applied to review and report an IRB approved, HIPAA compliant multiple-reader multiple-case clinical study of 150 bi-parametric prostate MRI studies across 9 readers, measuring physician performance both with and without the use of the recently FDA cleared CADe/CADx software ProstatID.

Results

Unassisted reader AUC values ranged from 0.418 – 0.759, with AI assisted AUC values ranging from 0.507 – 0.787. This represented a statistically significant AUC improvement of 0.045 (α = 0.05). A free-response ROC (FROC) analysis similarly demonstrated a statistically significant increase in θ from 0.405 to 0.453 (α = 0.05). The standalone performance of ProstatID performed across all prostate tissues demonstrated an AUC of 0.929, while the standalone lesion level performance of ProstatID at all biopsied locations achieved an AUC of 0.710.

Conclusion

This study applies and illustrates suggested reporting and standardization methods for prostate AI studies that will make it easier to understand, evaluate and compare between AI studies. Providing radiologists with the ProstatID CADe/CADx software significantly increased diagnostic performance as assessed by both ROC and free-response ROC metrics. Such algorithms have the potential to improve radiologist performance in the detection and localization of clinically significant prostate cancer.

MRI has firmly established itself as a mainstay for the detection, staging and surveillance of prostate cancer. Despite its success, prostate MRI continues to suffer from poor inter-reader variability and a low positive predictive value. The recent emergence of Artificial Intelligence (AI) to potentially improve diagnostic performance shows great potential. Understanding and interpreting the AI landscape as well as ever-increasing research literature, however, is difficult. This is in part due to widely varying study design and reporting techniques. This paper aims to address this need by first outlining the different types of AI used for the detection and diagnosis of prostate cancer, next deciphering how data collection methods, statistical analysis metrics (such as ROC and FROC analysis) and end points/outcomes (lesion detection vs. case diagnosis) affect the performance and limit the ability to compare between studies. Finally, this work explores the need for appropriately enriched investigational datasets and proper ground truth, and provides guidance on how to best conduct AI prostate MRI studies. Published in parallel, a clinical study applying this suggested study design was applied to review and report a multiple-reader multiple-case clinical study of 150 bi-parametric prostate MRI studies across nine readers, measuring physician performance both with and without the use of a recently FDA cleared Artificial Intelligence software.¹

There has been recent scrutiny of private equity involvement in the healthcare market by federal and state governmental agencies who are concerned about the corporatization and financialization of healthcare in the United States. Data is emerging that patient costs increase, quality of healthcare decreases, physician autonomy decreases, and physician burnout and moral injury increases when corporate interests like private equity enter the medical market. Like other medical specialties, the field of radiology has been affected by corporatization and radiologists should understand how private equity interests may affect individual radiologists and the radiology workforce on a larger scale.

Background

Non-OR Anesthesia (NORA) is rapidly becoming standard in many high-volume institutions and efficiency in these spaces has yet to be optimized. On-time first start percentage has been suggested to correlate with more efficient flow, and this correlation is established within the surgical space.

Purpose

To investigate the effects of timetable targets on first case on-time first start percentage within a NORA setting.

Materials and Methods

A retrospective study of anesthesia-supported first start cases from October 2022 to April 2023 was performed to analyze the effect of timetable targets on on-time first-case starts for planned cases. Statistical analysis was calculated using Student's t-tests with statistical significance defined as p < 0.05. Additionally, analysis of variance was used to compare three or more groups, and Tukey Kramer was used to evaluate groups pairwise.

Results

One hundred twenty-four first start cases were included in the evaluation. After intervention with timetable targets, average patient arrival to the room time improved from 7:49 AM to 7:40 AM (p < 0.05) and procedure start time improved from 8:31 AM to 8:20 AM (p < 0.01). The percentage of procedure start times occurring prior to the goal time increased from 35 % to 58 % after the implementation (p < 0.05). With exception of Tuesdays (Anesthesia Late Start Day), on-time starts improved from 17 % to 48 % (p < 0.01) and sustained this improvement throughout the post-implementation period.

Conclusion

Implementation of novel timetable targets yielded statistically significant improvement in first case start times. This improvement in efficiency and throughput results in increased room utilization, improved case throughput, and decreased block overrun times, all of which contribute toward increased revenues, decreased costs, and thus improved return on investment.

From mammographic screening guidelines to resident work hour regulations, public policy affects every aspect of the practice of radiology and ultimately determines how radiological care is delivered to patients. Shaping public policy through advocacy is therefore critical to ensure patient access to equitable, high-quality radiological care. In advocacy, individual practicing radiologists and radiology trainees can increase the scope of their influence by collaborating with professional radiology societies. When radiology trainees participate in organized radiology advocacy, they learn about regulatory and legislative issues that will affect their careers, and they learn how to effect policy change. Radiology societies in turn benefit from trainee involvement, as engaging trainees early in their careers leads to more robust future participation and leadership. To encourage trainee involvement, radiology societies can engage individual residency programs and medical student radiology interest groups, invest in trainee-focused events, and maximize the number of positions of responsibility open to trainees. To circumvent the barriers to participation that many trainees face, radiology societies can make meeting proceedings free and available through virtual mediums. Through active collaboration, trainees and professional societies can help assure a bright future for radiologists and patients in need of radiological care.

The shift from film to PACS in reading rooms, coupled with escalating case volumes, exposes radiologists to the issues of the modern computer workstation including computer work posture and work-related musculoskeletal disorders (WMSD). Common WMSDs affecting the neck and upper extremities include cervical myofascial pain, shoulder tendonitis, lateral epicondylitis, carpal tunnel syndrome, and cubital tunnel syndrome. This review examines each pathology along with its pathogenesis, clinical features, physical exam findings, and potential risk factors. Furthermore, a comprehensive 11-part physical therapy regimen that is both prophylactic and therapeutic is illustrated and described in detail. One of the objectives of this review is to advocate for the inclusion of a physical therapy regimen in the working routine of diagnostic radiologists to prevent WMSDs. A brief daily commitment to this regimen can help radiologists remain healthy and productive in order to deliver optimal patient care throughout their careers.

Social media are increasingly used as tools in radiologists education. This article describes features that aid with the selection of SM platforms, and how to emulate educator roles in the digital world. In addition, we summarize best practices regarding curating and delivering stellar content, building a SM brand, and rules of professionalism when using SM in radiology education.

Introduction

As opportunities for radiologists to subspecialize have increased, many avenues to organize Radiology department subspecialties exist. This study seeks to determine how academic U.S. Radiology departments structure themselves with respect to subspecialty divisions/sections, as there are no current standards for how Radiology departments are subdivided. Additionally, the extent of Radiology fellowships offered are assessed. The websites of academic U.S. Radiology departments, a highly influential source of information, were analyzed to perform this study.

Materials & methods

Radiology department websites of all allopathic U.S. medical schools (n = 148) were assessed for the following: presence/absence of Radiology department subdivisions, division/section labels, number of divisions/sections, division/section titles, presence/absence of Radiology fellowships, number of fellowships, and fellowships titles.

Results

114/148 (77 %) medical schools had Radiology department websites. According to their respective websites, 66/114 (58 %) academic Radiology departments had subspecialty divisions/sections, whereas 48/114 (42 %) had no divisions/sections listed. Of the departments that had divisions/sections, the median number of divisions/sections per department was nine, and ranged from two to 14. Fellowships were offered at 82/114 (72 %) academic Radiology departments that had websites, and the median number was six, ranging from one to 13.

Conclusion

There is marked heterogeneity of departmental organization across Radiology departments nationwide, likely due to the lack of current standards for how Radiology departments are subdivided into divisions/sections. Of the 77 % of medical schools that have Radiology department websites, only 58 % of departments listed divisions/sections, and 72 % posted fellowship offerings.

Objective

To determine if gadolinium-based contrast agents increase the sensitivity, specificity or reader confidence of malignant potential in musculoskeletal soft tissue tumors.

Methods

Pre- and post-contrast MRI studies from 87 patients were read by three independent radiologists of different experience. Readers noted malignant potential and confidence in their diagnosis based on pre-contrast and post-contrast MRI studies. Statistical models assessed for agreement between MRI reader diagnosis and pathologic results as well as analyzing effects of contrast on reader confidence. Inter- and intra-observer variabilities of malignant potential were also calculated.

Results

87 patients (48 benign and 39 malignant; mean [± SD] age 51 ± 17.9 and 57.1 ± 17.1, respectively) were evaluated. For all readers, pre-contrast and post-contrast sensitivities were 68.1 % and 70.6 % while pre-contrast and post-contrast specificities were 84.6 % and 83.8 %, respectively without significant change (p=0.88). There was not a significant association with the use of contrast and prediction of malignant potential with or without the resident reader (p=0.65 and p=0.82). Use of contrast was significantly associated with higher levels of reader confidence (p=0.02) for all readers. Inter- and intra-observer variabilities were in good agreement (W = 0.77 and 0.70).

Conclusion

The addition of a post-contrast sequence increased reader confidence in their diagnosis without a corresponding significant increase in accurate prediction of malignant potential.