Explore the vast range of titles available.

\"Implementation of artificial intelligence AI in Radiology is an important topic of discussion. Advances in AI which encompass machine learning, artificial neural networks, and deep learning are increasingly being applied to diagnostic imaging. While some posit radiologists are irreplaceable, certain AI proponents have proposed to stop training radiologists now. By compiling perspectives from experts from various backgrounds, this book explores the current state of AI efforts in Radiology along with the clinical, financial, technological, and societal perspectives on the role and expected impact of AI in Radiology\"-- Provided by publisher.

Accurate classification of focal liver lesions is an important part of liver disease diagnostics. In clinical practice, the lesion type is often determined from the abdominal MR examination, which includes T2-weighted and dynamic contrast enhanced (DCE) MR images. To date, only T2-weighted images are exploited for automatic classification of focal liver lesions. In this study additional MR sequences and risk factors are used for automatic classification to improve the results and to make a step forward to a clinically useful aid for radiologists. Clinical MRI data sets of 95 patients with in total 125 benign lesions (40 adenomas, 29 cysts and 56 hemangiomas) and 88 malignant lesions (30 hepatocellular carcinomas (HCC) and 58 metastases) were included in this study. Contrast curve, gray level histogram, and gray level co-occurrence matrix texture features were extracted from the DCE-MR and T2-weighted images. In addition, risk factors including the presence of steatosis, cirrhosis, and a known primary tumor were used as features. Fifty features with the highest ANOVA F-score were selected and fed to an extremely randomized trees classifier. The classifier evaluation was performed using the leave-one-out principle and receiver operating characteristic (ROC) curve analysis. The overall accuracy for the classification of the five major focal liver lesion types is 0.77. The sensitivity/specificity is 0.80/0.78, 0.93/0.93, 0.84/0.82, 0.73/0.56, and 0.62/0.77 for adenoma, cyst, hemangioma, HCC, and metastasis, respectively. The proposed classification system using features derived from clinical DCE-MR and T2-weighted images, with additional risk factors is able to differentiate five common types of lesions and is a step forward to a clinically useful aid for focal liver lesion diagnosis.

To evaluate whether incorporating CT perfusion imaging can significantly enhance diagnostic CT accuracy in stroke detection. Two 3rd-year residents (3rd of 5 years of residency) reviewed CT scans of 200 patients with suspected stroke, consisting of 104 patients with a proven stroke and a control group with 96 patients. They analyzed each patient in a blinded and randomized manner in two runs. In one session, they had only non-contrast CT and CT angiography available for diagnosis; in the other session at a later time point, an additional CT perfusion imaging was available. The performance achieved by the two readers was determined in terms of AUC (area under the curve), accuracy, sensitivity, specificity, positive and negative predictive value and Cohen’s Kappa. Reader 1 achieved an AUC of 87.64% with the basic stroke-protocol vs. an AUC of 97.4% with an additional CT-perfusion given. Based on the DeLong test, these values differ significantly (p-value: 0.00017). Reader 2 achieved an AUC of 91.23% in basic stroke-protocol vs. an AUC of 96.42% with an additional CT-perfusion. These values also differ significantly (p-value: 0.02612).. The performance gain achieved with CT-perfusion is most evident in the decrease in the number of false classified cases (Reader 1: 24 to 5; Reader 2: 18 or 14 to 7) and the significant increase in Cohen’s kappa. Our study shows that additional CT-perfusion imaging in stroke diagnosis significantly improves the diagnostic reliability of residents. Therefore, it should be further investigated whether perfusion imaging should be a general standard of initial stroke diagnosis no matter of the onset.

Objectives To assess the quality of simplified radiology reports generated with the large language model (LLM) ChatGPT and to discuss challenges and chances of ChatGPT-like LLMs for medical text simplification. Methods In this exploratory case study, a radiologist created three fictitious radiology reports which we simplified by prompting ChatGPT with “Explain this medical report to a child using simple language.” In a questionnaire, we tasked 15 radiologists to rate the quality of the simplified radiology reports with respect to their factual correctness, completeness, and potential harm for patients. We used Likert scale analysis and inductive free-text categorization to assess the quality of the simplified reports. Results Most radiologists agreed that the simplified reports were factually correct, complete, and not potentially harmful to the patient. Nevertheless, instances of incorrect statements, missed relevant medical information, and potentially harmful passages were reported. Conclusion While we see a need for further adaption to the medical field, the initial insights of this study indicate a tremendous potential in using LLMs like ChatGPT to improve patient-centered care in radiology and other medical domains. Clinical relevance statement Patients have started to use ChatGPT to simplify and explain their medical reports, which is expected to affect patient-doctor interaction. This phenomenon raises several opportunities and challenges for clinical routine. Key Points • Patients have started to use ChatGPT to simplify their medical reports, but their quality was unknown. • In a questionnaire, most participating radiologists overall asserted good quality to radiology reports simplified with ChatGPT. However, they also highlighted a notable presence of errors, potentially leading patients to draw harmful conclusions. • Large language models such as ChatGPT have vast potential to enhance patient-centered care in radiology and other medical domains. To realize this potential while minimizing harm, they need supervision by medical experts and adaption to the medical field. Graphical Abstract

ObjectivesTo assess undergraduate medical students’ attitudes towards artificial intelligence (AI) in radiology and medicine.Materials and methodsA web-based questionnaire was designed using SurveyMonkey, and was sent out to students at three major medical schools. It consisted of various sections aiming to evaluate the students’ prior knowledge of AI in radiology and beyond, as well as their attitude towards AI in radiology specifically and in medicine in general. Respondents’ anonymity was ensured.ResultsA total of 263 students (166 female, 94 male, median age 23 years) responded to the questionnaire. Around 52% were aware of the ongoing discussion about AI in radiology and 68% stated that they were unaware of the technologies involved. Respondents agreed that AI could potentially detect pathologies in radiological examinations (83%) but felt that AI would not be able to establish a definite diagnosis (56%). The majority agreed that AI will revolutionise and improve radiology (77% and 86%), while disagreeing with statements that human radiologists will be replaced (83%). Over two-thirds agreed on the need for AI to be included in medical training (71%). In sub-group analyses male and tech-savvy respondents were more confident on the benefits of AI and less fearful of these technologies.ConclusionContrary to anecdotes published in the media, undergraduate medical students do not worry that AI will replace human radiologists, and are aware of the potential applications and implications of AI on radiology and medicine. Radiology should take the lead in educating students about these emerging technologies.Key Points• Medical students are aware of the potential applications and implications of AI in radiology and medicine in general.• Medical students do not worry that the human radiologist or physician will be replaced.• Artificial intelligence should be included in medical training.

To investigate image quality and radiation dose of CT coronary angiography (CTCA) scanned using automatic tube current modulation (ATCM) and reconstructed by strong adaptive iterative dose reduction three-dimensional (AIDR3D). Eighty-four consecutive CTCA patients were collected for the study. All patients were scanned using ATCM and reconstructed with strong AIDR3D, standard AIDR3D and filtered back-projection (FBP) respectively. Two radiologists who were blinded to the patients' clinical data and reconstruction methods evaluated image quality. Quantitative image quality evaluation included image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). To evaluate image quality qualitatively, coronary artery is classified into 15 segments based on the modified guidelines of the American Heart Association. Qualitative image quality was evaluated using a 4-point scale. Radiation dose was calculated based on dose-length product. Compared with standard AIDR3D, strong AIDR3D had lower image noise, higher SNR and CNR, their differences were all statistically significant (P<0.05); compared with FBP, strong AIDR3D decreased image noise by 46.1%, increased SNR by 84.7%, and improved CNR by 82.2%, their differences were all statistically significant (P<0.05 or 0.001). Segments with diagnostic image quality for strong AIDR3D were 336 (100.0%), 486 (96.4%), and 394 (93.8%) in proximal, middle, and distal part respectively; whereas those for standard AIDR3D were 332 (98.8%), 472 (93.7%), 378 (90.0%), respectively; those for FBP were 217 (64.6%), 173 (34.3%), 114 (27.1%), respectively; total segments with diagnostic image quality in strong AIDR3D (1216, 96.5%) were higher than those of standard AIDR3D (1182, 93.8%) and FBP (504, 40.0%); the differences between strong AIDR3D and standard AIDR3D, strong AIDR3D and FBP were all statistically significant (P<0.05 or 0.001). The mean effective radiation dose was (2.55±1.21) mSv. Compared with standard AIDR3D and FBP, CTCA with ATCM and strong AIDR3D could significantly improve both quantitative and qualitative image quality.

BackgroundThe adequacy of leptomeningeal collateral flow has a pivotal role in determining clinical outcome in acute ischemic stroke. The American Society of Interventional and Therapeutic Neuroradiology/Society of Interventional Radiology (ASITN/SIR) collateral score is among the most commonly used scales for measuring this flow. It is based on the extent and rate of retrograde collateral flow to the impaired territory on angiography.ObjectiveTo evaluate inter- and intraobserver agreementin angiographic leptomeningeal collateral flow assessment.Materials and methodsThirty pretreatment angiogram video loops (frontal and lateral view), chosen from the randomized controlled trial THRombectomie des Artères CErebrales (THRACE), were sent for grading in an electronic file. 19 readers participated, including eight who had access to a training set before the first grading. 13 readers made a double evaluation, 3 months apart.ResultsOverall agreement among the 19 observers was poor (κ = 0,16 ± 6,5.10 -3), and not improved with prior training (κ = 0,14 ± 0,016). Grade 4 showed the poorest interobserver agreement (κ=0.18±0.002) while grades 0 and 1 were associated with the best results (κ=0.52±0.001 and κ=0.43±0.004, respectively). Interobserver agreement increased (κ = 0,27± 0,014) when a dichotomized score, ‘poor collaterals’ (score of 0, 1 or 2) versus ‘good collaterals’ (score of 3 or 4) was used. The intraobserver agreements varied between slight (κ=0.18±0.13) and substantial (κ=0.74±0.1), and were slightly improved with the dichotomized score (from κ=0.19±0.2 to κ=0.79±0.11).ConclusionInter- and intraobserver agreement of collateral circulation grading using the ASITN/SIR score was poor, raising concerns about comparisons among publications. A simplified dichotomized judgment may be a more reproducible assessment when images are rated by the same observer(s) in randomized trials.

Radiomics is a relatively new word for the field of radiology, meaning the extraction of a high number of quantitative features from medical images. Artificial intelligence (AI) is broadly a set of advanced computational algorithms that basically learn the patterns in the data provided to make predictions on unseen data sets. Radiomics can be coupled with AI because of its better capability of handling a massive amount of data compared with the traditional statistical methods. Together, the primary purpose of these fields is to extract and analyze as much and meaningful hidden quantitative data as possible to be used in decision support. Nowadays, both radiomics and AI have been getting attention for their remarkable success in various radiological tasks, which has been met with anxiety by most of the radiologists due to the fear of replacement by intelligent machines. Considering ever-developing advances in computational power and availability of large data sets, the marriage of humans and machines in future clinical practice seems inevitable. Therefore, regardless of their feelings, the radiologists should be familiar with these concepts. Our goal in this paper was three-fold: first, to familiarize radiologists with the radiomics and AI; second, to encourage the radiologists to get involved in these ever-developing fields; and, third, to provide a set of recommendations for good practice in design and assessment of future works.

The value of a computer-aided detection tool (CAD) as second reader in combination with experienced and inexperienced radiologists for the diagnosis of acute pulmonary embolism (PE) was assessed prospectively. Computed tomographic angiography (CTA) scans (64 × 0.6 mm collimation; 61.4 mm/rot table feed) of 56 patients (31 women, 34–89 years, mean = 66 years) with suspected PE were analysed by two experienced (R1, R2) and two inexperienced (R3, R4) radiologists for the presence and distribution of emboli using a five-point confidence rating, and by CAD. Informed consent was obtained from all patients. Results were compared with an independent reference standard. Inter-observer agreement was calculated by kappa, confidence assessed by ROC analysis. A total of 1,116 emboli [within mediastinal ( n  = 72), lobar ( n  = 133), segmental ( n  = 465) and subsegmental arteries ( n  = 455)] were included. CAD detected 343 emboli (sensitivity = 30.74%, correct-positive rate = 6.13/patient; false-positive rate = 4.1/patient). Inter-observer agreement was good (R1, R2: κ = 0.84, 95% CI = 0.81–0.87; R3, R4: κ = 0.79, 95% CI = 0.76–0.81). Extended inter-observer agreement was higher in mediastinal and lobar than in segmental and subsegmental arteries (κ = 0.84–0.86 and κ = 0.51–0.58 for mediastinal/lobar and segmental/subsegmental arteries, respectively P  < 0.05). Agreement between experienced and inexperienced readers was improved by CAD (κ = 0.60–0.62 and κ = 0.69–0.72 before and after CAD consensus, respectively P  < 0.05). The experienced outperformed the inexperienced readers (Az = 0.95, 0.93, 0.89 and 0.86 for R1–4, respectively, P  < 0.05). CAD significantly improved overall performances of readers 3 and 4 (Az = 0.86 for R3, R4 and Az = 0.89 for R3, R4 with CAD, P  < 0.05), by enhancing sensitivities in segmental/subsegmental arteries. CAD improved experienced readers’ sensitivities in segmental/subsegmental arteries (sens. = 0.93 and 0.90 for R1, R2 before and 0.97 and 0.94 for R1, R2 after CAD consensus, P  < 0.05), without significant improvement of their overall performances ( P  > 0.05). Particularly inexperienced readers benefit from consensus with CAD data, greatly improving detection of segmental and subsegmental emboli. This system is advocated as a second reader.

Newer imaging modalities such as CT and MR can provide more detailed information with thinner images and/or multiple series of images, and the time required to collect these images is shorter than before. [...]the number of images collected in each examination is increasing, whereas the number of radiologists who interpret these images is not. [...]deep learning models can also be used to alert radiologists and physicians to patients who require urgent treatment, as in the application described by Taylor and colleagues in the detection of pneumothorax [2]. [...]deep learning models trained to predict histopathological findings based on noninvasive images, such as the models described above that use MR to stage liver fibrosis [7], may help in reducing the risk of complications from invasive biopsy. [...]although a trained model may exhibit high performance in one task such as diagnosis of pneumonia, deep learning in its current forms cannot replace the radiologist’s role in detecting incidental findings such as asymptomatic tumors.