Choosing the right artificial intelligence solutions for your radiology department: key factors to consider.

The rapid evolution of artificial intelligence (AI), particularly in deep learning, has significantly impacted radiology, introducing an array of AI solutions for interpretative tasks. This paper provides radiology departments with a practical guide for selecting and integrating AI solutions, focusing on interpretative tasks that require the active involvement of radiologists. Our approach is not to list available applications or review scientific evidence, as this information is readily available in previous studies; instead, we concentrate on the essential factors radiology departments must consider when choosing AI solutions. These factors include clinical relevance, performance and validation, implementation and integration, clinical usability, costs and return on investment, and regulations, security, and privacy. We illustrate each factor with hypothetical scenarios to provide a clearer understanding and practical relevance. Through our experience and literature review, we provide insights and a practical roadmap for radiologists to navigate the complex landscape of AI in radiology. We aim to assist in making informed decisions that enhance diagnostic precision, improve patient outcomes, and streamline workflows, thus contributing to the advancement of radiological practices and patient care.

Diagnostic capabilities of artificial intelligence as an additional reader in a breast cancer screening program.

OBJECTIVES: We aimed to evaluate the early-detection capabilities of AI in a screening program over its duration, with a specific focus on the detection of interval cancers, the early detection of cancers with the assistance of AI from prior visits, and its impact on workload for various reading scenarios. MATERIALS AND METHODS: The study included 22,621 mammograms of 8825 women within a 10-year biennial two-reader screening program. The statistical analysis focused on 5136 mammograms from 4282 women due to data retrieval issues, among whom 105 were diagnosed with breast cancer. The AI software assigned scores from 1 to 100. Histopathology results determined the ground truth, and Youden's index was used to establish a threshold. Tumor characteristics were analyzed with ANOVA and chi-squared test, and different workflow scenarios were evaluated using bootstrapping. RESULTS: The AI software achieved an AUC of 89.6% (86.1-93.2%, 95% CI). The optimal threshold was 30.44, yielding 72.38% sensitivity and 92.86% specificity. Initially, AI identified 57 screening-detected cancers (83.82%), 15 interval cancers (51.72%), and 4 missed cancers (50%). AI as a second reader could have led to earlier diagnosis in 24 patients (average 29.92 ± 19.67 months earlier). No significant differences were found in cancer-characteristics groups. A hybrid triage workflow scenario showed a potential 69.5% reduction in workload and a 30.5% increase in accuracy. CONCLUSION: This AI system exhibits high sensitivity and specificity in screening mammograms, effectively identifying interval and missed cancers and identifying 23% of cancers earlier in prior mammograms. Adopting AI as a triage mechanism has the potential to reduce workload by nearly 70%. CLINICAL RELEVANCE STATEMENT: The study proposes a more efficient method for screening programs, both in terms of workload and accuracy. KEY POINTS: • Incorporating AI as a triage tool in screening workflow improves sensitivity (72.38%) and specificity (92.86%), enhancing detection rates for interval and missed cancers. • AI-assisted triaging is effective in differentiating low and high-risk cases, reduces radiologist workload, and potentially enables broader screening coverage. • AI has the potential to facilitate early diagnosis compared to human reading.

Quantitative evaluation of Saliency-Based Explainable artificial intelligence (XAI) methods in Deep Learning-Based mammogram analysis.

BACKGROUND: Explainable Artificial Intelligence (XAI) is prominent in the diagnostics of opaque deep learning (DL) models, especially in medical imaging. Saliency methods are commonly used, yet there's a lack of quantitative evidence regarding their performance. OBJECTIVES: To quantitatively evaluate the performance of widely utilized saliency XAI methods in the task of breast cancer detection on mammograms. METHODS: Three radiologists drew ground-truth boxes on a balanced mammogram dataset of women (n = 1496 cancer-positive and negative scans) from three centers. A modified, pre-trained DL model was employed for breast cancer detection, using MLO and CC images. Saliency XAI methods, including Gradient-weighted Class Activation Mapping (Grad-CAM), Grad-CAM++, and Eigen-CAM, were evaluated. We utilized the Pointing Game to assess these methods, determining if the maximum value of a saliency map aligned with the bounding boxes, representing the ratio of correctly identified lesions among all cancer patients, with a value ranging from 0 to 1. RESULTS: The development sample included 2,244 women (75%), with the remaining 748 women (25%) in the testing set for unbiased XAI evaluation. The model's recall, precision, accuracy, and F1-Score in identifying cancer in the testing set were 69%, 88%, 80%, and 0.77, respectively. The Pointing Game Scores for Grad-CAM, Grad-CAM++, and Eigen-CAM were 0.41, 0.30, and 0.35 in women with cancer and marginally increased to 0.41, 0.31, and 0.36 when considering only true-positive samples. CONCLUSIONS: While saliency-based methods provide some degree of explainability, they frequently fall short in delineating how DL models arrive at decisions in a considerable number of instances.

MRI only detected lesions: Can contrast enhanced mammography guided biopsy be an alternative method: Initial clinical findings.

OBJECTIVE: This study aims to analyze our initial findings regarding CEM-guided stereotactic vacuum-assisted biopsy for MRI-only detected lesions and compare biopsy times by MRI-guided biopsy. MATERIALS AND METHODS: In this retrospective analysis, CEM-guided biopsies of MRI-only detected breast lesions from December 2021 to June 2023were included. Patient demographics, breast density, lesion size, background parenchymal enhancement on CEM, lesion positioning, procedure duration, and number of scout views were documented. Initially, seven patients had CEM imaging before biopsy; for later cases, CEM scout views were used for simultaneous lesion depiction and targeting. RESULTS: Two cases were excluded from the initial 28 patients with 29 lesions resulting in a total of 27 lesions in 26 women (mean age:44.96 years). Lesion sizes ranged from 4.5 to 41 mm, with two as masses and the remaining as non-mass enhancements. Histopathological results identified nine malignancies (33.3 %, 9/27), including invasive cancers (55.6 %, 5/9) and DCIS (44.4 %, 4/9). The biopsy PPV rate was 33.3 %. Benign lesions comprised 66.7 %, with 22.2 % high-risk lesions. The biopsy success rate was 93.1 % (27/29), and minor complications occurred in seven cases (25.9 %, 7/27), mainly small hematomas and one vasovagal reaction (3.7 %, 1/27). Median number of scout views required was 2, with no significant differences between cases with or without prior CEM (P = 0.8). Median duration time for biopsy was 14 min, significantly shorter than MRI-guided bx at the same institution (P < 0.001) by 24 min with predominantly upright positioning of the patient (88.9 %) and horizontal approach of the needle (92.6 %). CONCLUSION: This study showed that CEM-guided biopsy is a feasible and safe alternative method and a faster solution for MRI-only detected enhancing lesions and can be accurately performed without the need for prior CEM imaging.

Thyroid hormones and ovarian reserve: a comprehensive study of women seeking infertility care.

BACKGROUND: Ovarian reserve is the number of oocytes remaining in the ovary and is one of the most important aspects of a woman's reproductive potential. Research on the association between thyroid dysfunction and ovarian reserve has yielded controversial results. In our study, we aimed to investigate the relationship between thyroid-stimulating hormone (TSH) levels and ovarian reserve markers. METHODS: From 1443 women seeking infertility care, the data of 1396 women aged between 20-45 years old who had a body mass index between 18-30 kg/m2 were recruited for this retrospective study. The anti-Müllerian hormone (AMH) and TSH relationship was analyzed with generalized linear and polynomial regression. RESULTS: Median age, follicle-stimulating hormone (FSH), AMH, and TSH levels were 36.79 years, 9.55 IU/L, 3.57 pmol/L, and 1.80 mIU/L, respectively. Differences between TSH groups were statistically significant in terms of AMH level, antral follicle count (AFC), and age (p = 0.007 and p = 0.038, respectively). A generalized linear regression model could not explain age-matched TSH levels concerning AMH levels (p > 0.05). TSH levels were utilized in polynomial regression models of AMH, and the 2nd degree was found to have the best fit. The inflection point of the model was 2.88 mIU/L. CONCLUSIONS: Our study shows a correlation between TSH and AMH values in a population of infertile women. Our results are as follows: a TSH value of 2.88 mIU/L yields the highest AMH result. It was also found that AMH and AFC were positively correlated, while AMH and FSH were negatively correlated.

Anatomically guided self-adapting deep neural network for clinically significant prostate cancer detection on bi-parametric MRI: a multi-center study.

OBJECTIVE: To evaluate the effectiveness of a self-adapting deep network, trained on large-scale bi-parametric MRI data, in detecting clinically significant prostate cancer (csPCa) in external multi-center data from men of diverse demographics; to investigate the advantages of transfer learning. METHODS: We used two samples: (i) Publicly available multi-center and multi-vendor Prostate Imaging: Cancer AI (PI-CAI) training data, consisting of 1500 bi-parametric MRI scans, along with its unseen validation and testing samples; (ii) In-house multi-center testing and transfer learning data, comprising 1036 and 200 bi-parametric MRI scans. We trained a self-adapting 3D nnU-Net model using probabilistic prostate masks on the PI-CAI data and evaluated its performance on the hidden validation and testing samples and the in-house data with and without transfer learning. We used the area under the receiver operating characteristic (AUROC) curve to evaluate patient-level performance in detecting csPCa. RESULTS: The PI-CAI training data had 425 scans with csPCa, while the in-house testing and fine-tuning data had 288 and 50 scans with csPCa, respectively. The nnU-Net model achieved an AUROC of 0.888 and 0.889 on the hidden validation and testing data. The model performed with an AUROC of 0.886 on the in-house testing data, with a slight decrease in performance to 0.870 using transfer learning. CONCLUSIONS: The state-of-the-art deep learning method using prostate masks trained on large-scale bi-parametric MRI data provides high performance in detecting csPCa in internal and external testing data with different characteristics, demonstrating the robustness and generalizability of deep learning within and across datasets. CLINICAL RELEVANCE STATEMENT: A self-adapting deep network, utilizing prostate masks and trained on large-scale bi-parametric MRI data, is effective in accurately detecting clinically significant prostate cancer across diverse datasets, highlighting the potential of deep learning methods for improving prostate cancer detection in clinical practice.

Inter-reader agreement of the prostate imaging quality (PI-QUAL) score for basic readers in prostate MRI: A multi-center study.

BACKGROUND: The Prostate Imaging Quality (PI-QUAL) score is the first step toward image quality assessment in multi-parametric prostate MRI (mpMRI). Previous studies have demonstrated moderate to excellent inter-rater agreement among expert readers; however, there is a need for studies to assess the inter-reader agreement of PI-QUAL scoring in basic prostate readers. OBJECTIVES: To assess the inter-reader agreement of the PI-QUAL score amongst basic prostate readers on multi-center prostate mpMRI. METHODS: Five basic prostate readers from different centers assessed the PI-QUAL scores independently using T2-weighted images, diffusion-weighted imaging (DWI) including apparent diffusion coefficient (ADC) maps, and dynamic-contrast-enhanced (DCE) images on mpMRI data obtained from five different centers following Prostate Imaging-Reporting and Data System Version 2.1. The inter-reader agreements amongst radiologists for PI-QUAL were evaluated using weighted Cohen's kappa. Further, the absolute agreements in assessing the diagnostic adequacy of each mpMRI sequence were calculated. RESULTS: A total of 355 men with a median age of 71 years (IQR, 60-78) were enrolled in the study. The pair-wise kappa scores ranged from 0.656 to 0.786 for the PI-QUAL scores, indicating good inter-reader agreements between the readers. The pair-wise absolute agreements ranged from 0.75 to 0.88 for T2W imaging, from 0.74 to 0.83 for the ADC maps, and from 0.77 to 0.86 for DCE images. CONCLUSIONS: Basic prostate radiologists from different institutions provided good inter-reader agreements on multi-center data for the PI-QUAL scores.

Deep learning for assessing image quality in bi-parametric prostate MRI: A feasibility study.

BACKGROUND: Although systems such as Prostate Imaging Quality (PI-QUAL) have been proposed for quality assessment, visual evaluations by human readers remain somewhat inconsistent, particularly among less-experienced readers. OBJECTIVES: To assess the feasibility of deep learning (DL) for the automated assessment of image quality in bi-parametric MRI scans and compare its performance to that of less-experienced readers. METHODS: We used bi-parametric prostate MRI scans from the PI-CAI dataset in this study. A 3-point Likert scale, consisting of poor, moderate, and excellent, was utilized for assessing image quality. Three expert readers established the ground-truth labels for the development (500) and testing sets (100). We trained a 3D DL model on the development set using probabilistic prostate masks and an ordinal loss function. Four less-experienced readers scored the testing set for performance comparison. RESULTS: The kappa scores between the DL model and the expert consensus for T2W images and ADC maps were 0.42 and 0.61, representing moderate and good levels of agreement. The kappa scores between the less-experienced readers and the expert consensus for T2W images and ADC maps ranged from 0.39 to 0.56 (fair to moderate) and from 0.39 to 0.62 (fair to good). CONCLUSIONS: Deep learning (DL) can offer performance comparable to that of less-experienced readers when assessing image quality in bi-parametric prostate MRI, making it a viable option for an automated quality assessment tool. We suggest that DL models trained on more representative datasets, annotated by a larger group of experts, could yield reliable image quality assessment and potentially substitute or assist visual evaluations by human readers.

Automated LVO detection and collateral scoring on CTA using a 3D self-configuring object detection network: a multi-center study.

The use of deep learning (DL) techniques for automated diagnosis of large vessel occlusion (LVO) and collateral scoring on computed tomography angiography (CTA) is gaining attention. In this study, a state-of-the-art self-configuring object detection network called nnDetection was used to detect LVO and assess collateralization on CTA scans using a multi-task 3D object detection approach. The model was trained on single-phase CTA scans of 2425 patients at five centers, and its performance was evaluated on an external test set of 345 patients from another center. Ground-truth labels for the presence of LVO and collateral scores were provided by three radiologists. The nnDetection model achieved a diagnostic accuracy of 98.26% (95% CI 96.25-99.36%) in identifying LVO, correctly classifying 339 out of 345 CTA scans in the external test set. The DL-based collateral scores had a kappa of 0.80, indicating good agreement with the consensus of the radiologists. These results demonstrate that the self-configuring 3D nnDetection model can accurately detect LVO on single-phase CTA scans and provide semi-quantitative collateral scores, offering a comprehensive approach for automated stroke diagnostics in patients with LVO.

Does deep learning software improve the consistency and performance of radiologists with various levels of experience in assessing bi-parametric prostate MRI?

OBJECTIVE: To investigate whether commercially available deep learning (DL) software improves the Prostate Imaging-Reporting and Data System (PI-RADS) scoring consistency on bi-parametric MRI among radiologists with various levels of experience; to assess whether the DL software improves the performance of the radiologists in identifying clinically significant prostate cancer (csPCa). METHODS: We retrospectively enrolled consecutive men who underwent bi-parametric prostate MRI at a 3 T scanner due to suspicion of PCa. Four radiologists with 2, 3, 5, and > 20 years of experience evaluated the bi-parametric prostate MRI scans with and without the DL software. Whole-mount pathology or MRI/ultrasound fusion-guided biopsy was the reference. The area under the receiver operating curve (AUROC) was calculated for each radiologist with and without the DL software and compared using De Long's test. In addition, the inter-rater agreement was investigated using kappa statistics. RESULTS: In all, 153 men with a mean age of 63.59 ± 7.56 years (range 53-80) were enrolled in the study. In the study sample, 45 men (29.80%) had clinically significant PCa. During the reading with the DL software, the radiologists changed their initial scores in 1/153 (0.65%), 2/153 (1.3%), 0/153 (0%), and 3/153 (1.9%) of the patients, yielding no significant increase in the AUROC (p > 0.05). Fleiss' kappa scores among the radiologists were 0.39 and 0.40 with and without the DL software (p = 0.56). CONCLUSIONS: The commercially available DL software does not increase the consistency of the bi-parametric PI-RADS scoring or csPCa detection performance of radiologists with varying levels of experience.

The first stop of transit migrant students on the road to Europe and America: The Turkish education system.

There have been forced migrations from one country to another due to wars, conflicts, and various reasons within and among countries. Turkey is a transit point from Asia to Europe. Accordingly, this study sheds light on experiences the refugee students have in the Turkish Education System. Therefore, this study is meant to be based on a qualitative approach as a case study method. The data were obtained through a semi-structured interview and a focus group interview. The information was gathered from nine students-three Syrians, two Iranians, two Iraqis, and two Afghans-as well as five teachers. According to the results, refugee students had difficulty learning and using a new language, adapting to a new country and environment, and getting used to new friends they made and teachers they were taught by. The results also indicate that students were concerned about their inability to travel to the United States and European countries.

A Comparative Study of Multiparametric MRI Sequences in Measuring Prostate Cancer Index Lesion Volume.

Objectives: To compare the effectiveness of individual multiparametric prostate MRI (mpMRI) sequences-T2W, diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC), and dynamic contrast-enhanced (DCE)-in assessing prostate cancer (PCa) index lesion volume using whole-mount pathology as the ground-truth; to assess the impact of an endorectal coil (ERC) on the measurements. Materials and Methods: We retrospectively enrolled 72 PCa patients who underwent 3T mpMRI with (n = 39) or without (n = 33) an ERC. A pathologist drew the index lesion borders on whole-mount pathology using planimetry (whole-mountvol). A radiologist drew the borders of the index lesion on each mpMRI sequence-T2Wvol, DWIvol, ADCvol, and DCEvol. Additionally, we calculated the maximum index lesion volume for each patient (maxMRIvol). The correlation and differences between mpMRI and whole-mount pathology in measuring the index lesion volume and the impact of an ERC were investigated. Results: The median T2Wvol, DWIvol, ADCvol, DCEvol, and maxMRIvol were 0.68 cm3, 0.97 cm3, 0.98 cm3, 0.82 cm3, and 1.13 cm3. There were good positive correlations between whole-mountvol and mpMRI sequences. However, all mpMRI-derived volumes underestimated the median whole-mountvol volume of 1.97 cm3 (P ≤ 0.001), with T2Wvol having the largest volumetric underestimation while DWIvol and ADCvol having the smallest. The mean relative index lesion volume underestimations of maxMRIvol were 39.16% ± 32.58% and 7.65% ± 51.91% with and without an ERC (P = 0.002). Conclusion: T2Wvol, DWIvol, ADCvol, DCEvol, and maxMRIvol substantially underestimate PCa index lesion volume compared with whole-mount pathology, with T2Wvol having the largest volume underestimation. Additionally, using an ERC exacerbates the volume underestimation.

Diagnostic Performance of AI for Cancers Registered in A Mammography Screening Program: A Retrospective Analysis.

Purpose: To evaluate the performance of an artificial intelligence (AI) algorithm in a simulated screening setting and its effectiveness in detecting missed and interval cancers. Methods: Digital mammograms were collected from Bahcesehir Mammographic Screening Program which is the first organized, population-based, 10-year (2009-2019) screening program in Turkey. In total, 211 mammograms were extracted from the archive of the screening program in this retrospective study. One hundred ten of them were diagnosed as breast cancer (74 screen-detected, 27 interval, 9 missed), 101 of them were negative mammograms with a follow-up for at least 24 months. Cancer detection rates of radiologists in the screening program were compared with an AI system. Three different mammography assessment methods were used: (1) 2 radiologists' assessment at screening center, (2) AI assessment based on the established risk score threshold, (3) a hypothetical radiologist and AI team-up in which AI was considered to be the third reader. Results: Area under curve was 0.853 (95% CI = 0.801-0.905) and the cut-off value for risk score was 34.5% with a sensitivity of 72.8% and a specificity of 88.3% for AI cancer detection in ROC analysis. Cancer detection rates were 67.3% for radiologists, 72.7% for AI, and 83.6% for radiologist and AI team-up. AI detected 72.7% of all cancers on its own, of which 77.5% were screen-detected, 15% were interval cancers, and 7.5% were missed cancers. Conclusion: AI may potentially enhance the capacity of breast cancer screening programs by increasing cancer detection rates and decreasing false-negative evaluations.