Machine learning approaches in the prediction of positive axillary lymph nodes post neoadjuvant chemotherapy using MRI, CT, or ultrasound: A systematic review

Background and objective

Neoadjuvant chemotherapy is a standard treatment approach for locally advanced breast cancer. Conventional imaging modalities, such as magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound, have been used for axillary lymph node evaluation which is crucial for treatment planning and prognostication. This systematic review aims to comprehensively examine the current research on applying machine learning algorithms for predicting positive axillary lymph nodes following neoadjuvant chemotherapy utilizing imaging modalities, including MRI, CT, and ultrasound.

Methods

A systematic search was conducted across databases, including PubMed, Scopus, and Web of Science, to identify relevant studies published up to December 2023. Articles employing machine learning algorithms to predict positive axillary lymph nodes using MRI, CT, or ultrasound data after neoadjuvant chemotherapy were included. The review follows the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, encompassing data extraction and quality assessment.

Results

Seven studies were included, comprising 1502 patients. Four studies used MRI, two used CT, and one applied ultrasound. Two studies developed deep-learning models, while five used classic machine-learning models mainly based on multiple regression. Across the studies, the models showed high predictive accuracy, with the best-performing models combining radiomics and clinical data.

Conclusion

This systematic review demonstrated the potential of utilizing advanced data analysis techniques, such as deep learning radiomics, in improving the prediction of positive axillary lymph nodes in breast cancer patients following neoadjuvant chemotherapy.

Diagnostic performance of different imaging modalities for splenic malignancies: A comparative meta-analysis

Background and objectives

The spleen hosts both benign and malignant lesions. Despite multiple imaging modalities, the distinction between these lesions poses a diagnostic challenge, marked by varying diagnostic accuracy levels across methods. In this study, we aimed to evaluate and compare the diagnostic performance of various imaging techniques for detecting malignant splenic lesions.

Methods

Following PRISMA guidelines, we searched PubMed, Scopus, and Web of Sciences databases for studies evaluating imaging techniques in detecting malignant splenic lesions. Data extraction included diagnostic accuracy metrics, and methodological quality was assessed using QUADAS-2. Diagnostic Test Accuracy meta-analyses were conducted using R (version: 4.2.1). Subgroup analyses and meta-regression were performed to compare different modalities and clinical settings.

Results

Our study included 28 studies (pooled sample size: 2358), primarily using retrospective designs with histopathology as the reference standard. PET scan demonstrated the highest diagnostic accuracy (AUC: 92 %), demonstrating a sensitivity of 93 % (95 % CI: 80.4 % - 97.7 %) and a specificity of 82.8 % (95 % CI: 71.1 % - 90.4 %). Contrast-enhanced ultrasound (CEUS), Contrast-enhanced CT scan, and contrast-enhanced MRI also showed impressive performance with AUCs of 91.4 %, 90.9 %, and 85.3 %, respectively. Differences among these modalities were not statistically significant, but they outperformed non-contrast-enhanced methods. PET and CEUS exhibited higher specificity for lymphoma cases compared to studies including other malignancies.

Conclusion and clinical implications

Overall, PET emerges as the best modality for splenic malignancies, and CEUS and CE-MRI show promise as potential alternatives, notably due to their reduced radiation exposure. Further research is essential for precise malignancy differentiation.

Ultrasound and magnetic resonance imaging features of fetal urogenital anomalies: A pictorial essay

This pictorial essay focuses on ultrasound (US) and magnetic resonance imaging (MRI) features of fetal urogenital anomalies. Fetal urogenital malformations account for 30%-50% of all anomalies discovered during pregnancy or at birth. They are usually detected by fetal ultrasound exams. However, when ultrasound data on their characteristics is insufficient, MRI is the best option for detecting other associated anomalies. The prognosis highly depends on their type and whether they are associated with other fetal abnormalities.

Quantitative analysis of lung lesions using unenhanced chest computed tomography images

INTRODUCTION

Chest radiograph and computed tomography (CT) scans can accidentally reveal pulmonary nodules. Malignant and benign pulmonary nodules can be difficult to distinguish without specific imaging features, such as calcification, necrosis, and contrast enhancement. However, these lesions may exhibit different image texture characteristics which cannot be assessed visually. Thus, a computer-assisted quantitative method like histogram analysis (HA) of Hounsfield unit (HU) values can improve diagnostic accuracy, reducing the need for invasive biopsy.

METHODS

In this exploratory control study, nonenhanced chest CT images of 20 patients with benign (10) and cancerous (10) lesion were selected retrospectively. The appearances of benign and malignant lesions were very similar in chest CT images, and only pathology report was used to discriminate them. Free hand region of interest (ROI) was inserted inside the lesion for all slices of each lesion. Mean, minimum, maximum, and standard deviations of HU values were recorded and used to make HA.

RESULTS

HA showed that the most malignant lesions have a mean HU value between 30 and 50, a maximum HU less than 150, and a minimum HU between -30 and 20. Lesions outside these ranges were mostly benign.

CONCLUSION

Quantitative CT analysis may differentiate malignant from benign lesions without specific malignancy patterns on unenhanced chest CT image.

A meta-analysis on the diagnostic utility of ultrasound in pediatric distal forearm fractures

Pediatric distal forearm fractures, comprising 30% of musculoskeletal injuries in children, are conventionally diagnosed using radiography. Ultrasound has emerged as a safer diagnostic tool, eliminating ionizing radiation, enabling bedside examinations with real-time imaging, and proving effective in non-hospital settings. The objective of this study is to evaluate the diagnostic efficacy of ultrasound for detecting distal forearm fractures in the pediatric population. A systematic review and meta-analysis were conducted through a comprehensive literature search in PubMed, Scopus, Web of Science, and Embase databases until October 1, 2023, following established guidelines. Eligible studies, reporting diagnostic accuracy measures of ultrasound in pediatric patients with distal forearm fractures, were included. Relevant data elements were extracted, and data analysis was performed. The analysis included 14 studies with 1377 patients, revealing pooled sensitivity and specificity of 94.5 (95% CI 92.7-95.9) and 93.5 (95% CI 89.6-96.0), respectively. Considering pre-test probabilities of 25%, 50%, and 75% for pediatric distal forearm fractures, positive post-test probabilities were 83%, 44%, and 98%, while negative post-test probabilities were 2%, 6%, and 15%, respectively. The bivariate model indicated significantly higher diagnostic accuracy in the subgroup with trained ultrasound performers vs. untrained performers (p = 0.03). Furthermore, diagnostic accuracy was significantly higher in the subgroup examining radius fractures vs. ulna fractures (p < 0.001), while no significant differences were observed between 4-view and 6-view ultrasound subgroups or between radiologist ultrasound interpreters and non-radiologist interpreters. This study highlighted ultrasound's reliability in detecting pediatric distal forearm fractures, emphasizing the crucial role of expertise in precisely confirming fractures through ultrasound examinations.

A Meta-analysis of Burnout in Radiology Trainees and Radiologists: Insights From the Maslach Burnout Inventory

RATIONALE AND OBJECTIVES

Burnout, stemming from chronic work stress, is a significant issue in the medical field, especially among radiologists. It leads to compromised patient care and reduced job satisfaction. Addressing burnout in radiology is essential for the well-being of radiologists and, in turn, for improving patient care. This study aimed to assess the prevalence and dimensions of burnout among radiology trainees (RTs) and practicing radiologists (PRs).

MATERIALS AND METHODS

A systematic review and meta-analysis were conducted in accordance with established guidelines. The search encompassed PubMed, Scopus, Web of Science, and Embase databases up to June 20, 2023. Eligible studies that assessed the rate of burnout syndrome and/or its subscales, including depersonalization (DP), emotional exhaustion (EE), and personal accomplishment (PA), among RTs and/or PRs using the Maslach Burnout Inventory (MBI), were included. Relevant data were extracted and analyzed using R and STATA.

RESULTS

Among the 22 included studies, the pooled rates of positive MBI subscales for RTs and PRs were as follows: 54.7% (95% confidence interval [CI]: 43.8%-65.1%, I2 = 95.2%) for DP, 57.2% (95% CI: 48.7%-65.4%, I2 = 92.9%) for EE, and 38.6% (95% CI: 27%-51.7%, I2 = 95.5%) for low PA. The pooled rate indicating the presence of at least one positive MBI subscale was 82.9% (95% CI: 79.2%-86.1%, I2 = 57.4%). For two or more positive MBI subscales, the pooled rate was 55.5% (95% CI: 49.7%-61.3%, I2 = 60.2%), and for three positive MBI subscales, it was 16.7% (95% CI: 11.7%-23.3%, I2 = 82.7%).

CONCLUSION

This study emphasizes a notable prevalence of burnout in the radiology specialty, with 8 of 10 individuals exhibiting positive results in at least one MBI subscale. This highlights the urgent need for interventions and support systems to protect the well-being of both trainees and practitioners and uphold the quality of patient care.

A systematic review and meta-analysis on the diagnostic utility of ultrasound for clavicle fractures

BACKGROUND

Ultrasound's real-time, high-resolution imaging, and accessibility at the point of care make it a valuable tool for diagnosing clavicle fractures, with growing evidence supporting its diagnostic accuracy compared to other imaging modalities.

OBJECTIVE

To assess the diagnostic utility of ultrasound in detecting clavicle fractures.

METHODS

A systematic review and meta-analysis were conducted by performing a comprehensive literature search in PubMed, Scopus, Web of Science, and Embase databases up to March 10, 2023, following established guidelines. Eligible studies that reported outcomes of interest were included, relevant data elements were extracted, and data analysis was performed using STATA software version 17.0.

RESULTS

Meta-analysis of seven included studies showed high-pooled sensitivity (0.94) and specificity (0.98) values for ultrasonography in diagnosing clavicle fractures, with low to moderate heterogeneity for sensitivity and high heterogeneity for specificity. Meta-regression and subgroup analyses revealed that pediatric studies had higher sensitivity but significantly lower specificity (P=0.01) compared to mixed or adult studies. Additional subgroup analysis in the pediatric group indicated decreased heterogeneity for specificity. Fagan plot analysis demonstrated favorable post-test probabilities for both positive and negative results across varying pre-test probabilities. Additionally, the likelihood ratio scatter matrix showed moderate to high test performance for both exclusion and confirmation purposes.

CONCLUSION

The current literature supports ultrasound as a reliable imaging modality for detecting clavicle fractures. It offers accurate diagnosis without exposing patients, especially children, to radiation.

Net water uptake as a predictive neuroimaging marker for acute ischemic stroke outcomes: a meta-analysis

OBJECTIVE

To assess the role of net water uptake (NWU) in predicting outcomes in acute ischemic stroke (AIS) patients.

METHODS

A systematic review and meta-analysis were performed, adhering to established guidelines. The search covered PubMed, Scopus, Web of Science, and Embase databases until July 1, 2023. Eligible studies reporting quantitative ischemic lesion NWU in admission CT scans of AIS patients, stratified based on outcomes, were included. Data analysis was performed using R software version 4.2.1.

RESULTS

Incorporating 17 original studies with 2217 AIS patients, NWU was significantly higher in patients with poor outcomes compared to those with good outcomes (difference of medians: 5.06, 95% CI: 3.00-7.13, p < 0.001). Despite excluding one outlier study, considerable heterogeneity persisted among the included studies (I2 = 90.8%). The meta-regression and subgroup meta-analyses demonstrated significantly higher NWU in patients with poor functional outcome, as assessed by modified Rankin Scale (difference of medians: 3.83, 95% CI: 1.98-5.68, p < 0.001, I2 = 72.9%), malignant edema/infarct (difference of medians: 8.30, 95% CI: 4.01-12.58, p < 0.001, I2 = 95.6%), and intracranial hemorrhage (difference of medians: 5.43, 95% CI: 0.44-10.43, p = 0.03, I2 = 91.1%).

CONCLUSION

NWU on admission CT scans shows promise as a predictive marker for outcomes in AIS patients. Prospective, multicenter trials with standardized, automated NWU measurement are crucial for robustly predicting diverse clinical outcomes.

CLINICAL RELEVANCE STATEMENT

The potential of net water uptake as a biomarker for predicting outcomes in acute ischemic stroke patients holds significant promise. Further validation through additional research could lead to its integration into clinical practice, potentially improving the accuracy of clinical decision-making and allowing for the development of more precise patient care strategies.

KEY POINTS

• Net water uptake, a CT-based biomarker, quantifies early brain edema after acute ischemic stroke. • Net water uptake is significantly higher in poor outcome acute ischemic stroke patients. • Net water uptake on CT scans holds promise in predicting diverse acute ischemic stroke outcomes.

Flow diverters in the treatment of intracranial dissecting aneurysms: a systematic review and meta-analysis of safety and efficacy

BACKGROUND

Intracranial dissecting aneurysms present clinical challenges, demanding immediate intervention due to their high bleeding risk. While traditional treatments have limitations, the potential of flow diverters shows promise but remains a subject of ongoing debate for optimal management. The aim of this study was assess the safety and efficacy of flow diverters in the treatment of intracranial dissecting aneurysms.

METHODS

A systematic review and meta-analysis was performed following established guidelines. The search encompassed PubMed, Scopus, Web of Science, and Embase databases up to July 20, 2023. Eligible studies reporting outcomes of interest were included, and relevant data were extracted and analyzed using R software.

RESULTS

The analysis, based on data pooled from 20 included studies involving 329 patients, revealed a favorable functional outcome rate of 89.7% at the last follow-up. The mortality rate during the follow-up period was 2.4%, decreasing to 0.9% when excluding the outlier study. In the final angiographic follow-up, a complete occlusion rate of 71.7% and an adequate occlusion rate of 88.3% were observed. Notably, studies with longer angiographic follow-up times exhibited lower rates of complete (P=0.02) and adequate (P<0.01) occlusion. A minimal aneurysm recurrence/rebleeding rate of 0.1% was noted, while in-stent stenosis/thrombosis occurred at a rate of 1.14%. Additionally, ischemic events/infarctions were seen in 3.3% of cases. The need for retreatment was minimal, with a rate of 0.9%, and the technical success rate was impressively high at 99.1%.

CONCLUSION

This study highlights the safety and efficacy of flow diverters in treating intracranial dissecting aneurysms. Further research, encompassing larger multicenter studies with extended follow-up periods, is crucial for comprehending occlusion dynamics, refining treatment strategies, improving long-term outcomes, and addressing methodological limitations.

Mechanical Thrombectomy for In-Hospital Onset Stroke: A Comparative Systematic Review and Meta-Analysis

BACKGROUND AND PURPOSE

In-hospital onset stroke (IHOS) accounts for a significant proportion of large vessel occlusion acute ischemic strokes, leading to worse outcomes due to delays in evaluation and treatment. Limited data is available on the effectiveness of mechanical thrombectomy in IHOS patients. This study aims to assess the safety and efficacy of mechanical thrombectomy for patients with IHOS and compare the outcomes with those of community-onset strokes (COS).

METHODS

We conducted a systematic review and meta-analysis following established guidelines, by searching PubMed, Scopus, Web of Science, and Embase databases up to April 11, 2023. Eligible studies reporting outcomes of interest were included, and relevant data was extracted and analyzed using Stata software version 17.0.

RESULTS

In a meta-analysis of nine studies, comprising 540 cases of IHOS and 5,744 cases of COS, IHOS cases had a significantly lower rate of good functional outcomes on follow-up (35.46% vs. 40.74%, P<0.01) and a higher follow-up mortality rate (26.29% vs. 18.08%, P<0.01) compared to COS patients. Both groups had comparable successful recanalization rates (IHOS: 79.32% vs. COS: 81.44%, P=0.11), incidence rates of periprocedural complications (IHOS: 15.10%, COS: 12.96%, P=0.78), and symptomatic intracranial hemorrhage (IHOS: 6.24%, COS: 6.88%, P=0.67). It is worth noting that much of the observed effect size for mortality and good functional outcomes on follow-up was derived from only one and two studies, respectively.

CONCLUSION

While the current literature suggests that mechanical thrombectomy is a safe and effective treatment for IHOS, further research is necessary to comprehensively evaluate its impact, particularly during follow-up.

Radiology as a Specialty in the Era of Artificial Intelligence: A Systematic Review and Meta-analysis on Medical Students, Radiology Trainees, and Radiologists

RATIONALE AND OBJECTIVES

Artificial intelligence (AI) is changing radiology by automating tasks and assisting in abnormality detection and understanding perceptions of medical students, radiology trainees, and radiologists is vital for preparing them for AI integration in radiology.

MATERIALS AND METHODS

A systematic review and meta-analysis were conducted following established guidelines. PubMed, Scopus, and Web of Science were searched up to March 5, 2023. Eligible studies reporting outcomes of interest were included, and relevant data were extracted and analyzed using STATA software version 17.0.

RESULTS

A meta-analysis of 21 studies revealed that 22.36% of individuals were less likely to choose radiology as a career due to concerns about advances in AI. Medical students showed higher rates of concern (31.94%) compared to radiology trainees and radiologists (9.16%) (P < .01). Radiology trainees and radiologists also demonstrated higher basic AI knowledge (71.84% vs 35.38%). Medical students had higher rates of belief that AI poses a threat to the radiology job market (42.66% vs 6.25%, P < .02). The pooled rate of respondents who believed that "AI will revolutionize radiology in the future" was 79.48%, with no significant differences based on participants' positions. The pooled rate of responders who believed in the integration of AI in medical curricula was 81.75% among radiology trainees and radiologists and 70.23% among medical students.

CONCLUSION

The study revealed growing concerns regarding the impact of AI in radiology, particularly among medical students, which can be addressed by revamping education, providing direct AI experience, addressing limitations, and emphasizing medico-legal issues to prepare for AI integration in radiology.

Disparities in computed tomography utilization for pediatric blunt trauma: a systematic review and meta-analysis comparing pediatric and non-pediatric trauma centers

Pediatric blunt trauma is a major cause of morbidity and mortality, and computed tomography (CT) imaging is vital for accurate evaluation and management. Pediatric trauma centers (PTCs) have selective CT practices, while non-PTCs may differ, resulting in potential variations in CT utilization. The objective of this study is to delineate disparities in CT utilization for pediatric blunt trauma patients between PTCs and non-PTCs. A systematic review and meta-analysis were conducted following established guidelines, searching PubMed, Scopus, and Web of Science up to March 3, 2023. All studies examining CT utilization in the management of pediatric (aged < 21 years) blunt trauma and specifying the type of trauma center(s) were included, and data were extracted and analyzed using STATA software version 17.0. An analysis of 30 studies revealed significant variations in CT scan utilization among pediatric blunt trauma patients across different types of trauma centers. PTCs exhibited lower pooled rates of abdominopelvic CT scans (35.4% vs. 44.9%, p < 0.01), cranial CT scans (36.9% vs. 42.9%, p < 0.01), chest CT scans (14.5% vs. 25.4%, p < 0.01), and cervical spine CT scans (23% vs. 45%, p < 0.01) compared to adult or mixed trauma centers (ATCs/MTCs). PTCs had a pooled rate of 54% for receiving at least one CT scan, while ATCs/MTCs had a higher rate of 69.3% (p < 0.05). The studies demonstrated considerable heterogeneity. These findings underscore the need to conduct further research to understand the reasons for the observed variations and to promote appropriate imaging usage, minimize radiation exposure, and encourage collaboration between pediatric and adult trauma centers.

Diagnostic utility of multidetector CT scan in penetrating diaphragmatic injuries: A systematic review and meta-analysis

Penetrating diaphragmatic injuries pose diagnostic and management challenges. Computed tomography (CT) scans are valuable for stable patients, but concern exists for missed injuries and complications in nonoperatively managed cases. The objective of this study was to explore the diagnostic utility of multidetector CT scan (MDCT) in identifying diaphragmatic injuries resulting from penetrating trauma. A systematic review and meta-analysis were conducted, following established guidelines, by searching PubMed, Scopus, Web of Science, and Embase databases up to July 6, 2023. Eligible studies reporting MDCT's diagnostic accuracy in detecting penetrating diaphragmatic injuries were included. Relevant data elements were extracted and analyzed using STATA software. The study included 9 articles comprising 294 patients with confirmed penetrating diaphragmatic injuries through surgical procedures. MDCT's diagnostic performance revealed a pooled sensitivity of 74% (95% CI: 56%-87%) and a pooled specificity of 92% (95% CI: 79%-97%) (Fig. two), with significant heterogeneity in both sensitivity and specificity across the studies. The Fagan plot demonstrated that higher pre-test probabilities correlated with higher positive post-test probabilities for penetrating diaphragmatic injury diagnosis using MDCT, but even with negative results, there remained a small chance of having the injury, especially in cases with higher pre-test probabilities. This study highlights MDCT's effectiveness in detecting diaphragmatic injury from penetrating trauma, with moderate to high diagnostic accuracy. However, larger sample sizes, multicenter collaborations, and prospective designs are needed to address observed heterogeneity, enhancing understanding and consistency in MDCT's diagnostic capabilities in this context.

A systematic review and meta-analysis of incidental findings in computed tomography scans for pediatric trauma patients

PURPOSE

To quantitatively synthesize and report the frequency and category of incidental findings on Computed Tomography (CT) scans in pediatric trauma patients.

METHODS

A thorough literature search was carried out in PubMed, Scopus, and Web of Science databases until March 6, 2023, in adherence to the preferred reporting items for systematic review and meta-analyses (PRISMA) guidelines. Studies describing incidental findings on CT scans in trauma patients ≤21 years were included. Incidental findings were grouped into three categories: Category 1 (requiring immediate or urgent evaluation or treatment), Category 2 (likely benign but which may require outpatient follow-up), and Category 3 (benign anatomic variants or pathologic findings that do not require follow-up or intervention).

RESULTS

Seven studies were included in this study, which revealed a combined rate of 27.10 % of incidental findings with notable heterogeneity among the studies. Aggregated frequencies were 10.15 % for Category 1, 32.18 % for Category 2 and 51.44 % for Category 3. Subgroup meta-analysis on abdominal CT scans showed a higher pooled incidence of incidental findings at 47.17 %, but with lower heterogeneity than the general meta-analysis.

CONCLUSION

The study underscores the prevalence of incidental findings in pediatric trauma patients undergoing CT scans. The categorization of these findings provides useful information for clinicians in determining appropriate follow-up and management strategies.