Search

Intracranial hemorrhages require an immediate diagnosis to optimize patient management and outcomes, and CT is the modality of choice in the emergency setting. We aimed to evaluate the performance of the first scanner-integrated artificial intelligence algorithm to detect brain hemorrhages in a routine clinical setting. This retrospective study includes 435 consecutive non-contrast head CT scans. Automatic brain hemorrhage detection was calculated as a separate reconstruction job in all cases. The radiological report (RR) was always conducted by a radiology resident and finalized by a senior radiologist. Additionally, a team of two radiologists reviewed the datasets retrospectively, taking additional information like the clinical record, course, and final diagnosis into account. This consensus reading served as a reference. Statistics were carried out for diagnostic accuracy. Brain hemorrhage detection was executed successfully in 432/435 (99%) of patient cases. The AI algorithm and reference standard were consistent in 392 (90.7%) cases. One false-negative case was identified within the 52 positive cases. However, 39 positive detections turned out to be false positives. The diagnostic performance was calculated as a sensitivity of 98.1%, specificity of 89.7%, positive predictive value of 56.7%, and negative predictive value (NPV) of 99.7%. The execution of scanner-integrated AI detection of brain hemorrhages is feasible and robust. The diagnostic accuracy has a high specificity and a very high negative predictive value and sensitivity. However, many false-positive findings resulted in a relatively moderate positive predictive value. [ABSTRACT FROM AUTHOR]

The aim of this study was to assess the diagnostic value of ADC distribution curves for differentiation between benign and malignant parotid gland tumors and to compare with mean ADC values. 73 patients with parotid gland tumors underwent head-and-neck MRI on a 1.5 Tesla scanner prior to surgery and histograms of ADC values were extracted. Histopathological results served as a reference standard for further analysis. ADC histograms were evaluated by comparing their similarity to a reference distribution using Chi2-test-statistics. The assumed reference distribution for benign and malignant parotid gland lesions was calculated after pooling the entire ADC data. In addition, mean ADC values were determined. For both methods, we calculated and compared the sensitivity and specificity between benign and malignant parotid gland tumors and three subgroups (pleomorphic adenoma, Warthin tumor, and malignant lesions), respectively. Moreover, we performed cross-validation (CV) techniques to estimate the predictive performance between ADC distributions and mean values. Histopathological results revealed 30 pleomorphic adenomas, 22 Warthin tumors, and 21 malignant tumors. ADC histogram distribution yielded a better specificity for detection of benign parotid gland lesions (ADChistogram: 75.0% vs. ADCmean: 71.2%), but mean ADC values provided a higher sensitivity (ADCmean: 71.4% vs. ADChistogram: 61.9%). The discrepancies are most pronounced in the differentiation between malignant and Warthin tumors (sensitivity ADCmean: 76.2% vs. ADChistogram: 61.9%; specificity ADChistogram: 81.8% vs. ADCmean: 68.2%). Using CV techniques, ADC distribution revealed consistently better accuracy to differentiate benign from malignant lesions ("leave-one-out CV" accuracy ADChistogram: 71.2% vs. ADCmean: 67.1%). ADC histogram analysis using full distribution curves is a promising new approach for differentiation between primary benign and malignant parotid gland tumors, especially with respect to the advantage in predictive performance based on CV techniques. [ABSTRACT FROM AUTHOR]

Pulmonary arterial dual-energy (aDE) CT is an established technique for evaluating pulmonary perfusion blood volume (PBV). As DECT protocols are increasingly used for thoraco-abdominal CT, this study assessed image quality and clinical findings in portal–venous phase dual-energy (vDE) CT and compared it to aDE. In 95 patients, vDE-CT was performed using a dual-source scanner (70/Sn150 kV, 560/140 ref.mAs). Pulmonary triggered aDE-CT served as reference (n = 94). PBV was reconstructed using a dedicated algorithm. Mean relative attenuation was measured in the pulmonary trunk, aorta, and segmented lung parenchyma. A distribution ratio (DL) between vessels and parenchyma was calculated to assess the iodine uptake of the lung parenchyma. Subjective overall diagnostic image quality was assessed for PBV images on a five-point Likert scale. Image artifacts were classified into five groups based on scale rating and compared between vDE and aDE. Pathological findings were correlated with the anatomical image datasets. Mean relative attenuation of the lung parenchyma was comparable in both groups (vDE: 23 ± 6 HU and aDE: 22 ± 7 HU), but significantly lower in the vessels of vDE. Therefore, iodine uptake of the lung parenchyma was significantly higher in vDE (DL: 10% vs. 8%, p < 0.01). The subjective overall image quality of the PBV images was comparable (p = 0.5). Rotation and streak artifacts were found in most of the patients (>86%, both p > 0.6). Dual-source artifacts were found in only a few patients in both groups (vDE 5%, aDE 7%, p = 0.5). Recess and subpleural artifacts were increased in vDE (vDE 53/27%, aDE 24/7%, both p < 0.001). Pathological findings were found in 19% of the vDE patients and 59% of the aDE patients. Comparable objective and subjective image quality of lung perfusion can be obtained in vDE and aDE. Iodine uptake of the lung parenchyma is increased in vDE compared to aDE, suggesting an interstitial pooling effect. Knowledge of the different appearances of artifacts will aid in the interpretation of the images. Additional clinical information about the lung parenchyma can be provided by PBV evaluation in vDE. [ABSTRACT FROM AUTHOR]

Low-dose dual-source computed tomography (DSCT) protocols for the evaluation of lung diseases in children and adolescents are of importance since this age group is particularly prone to radiation damage. The aim of this study was to evaluate image quality of low-dose DSCT of the lung and to assess the potential of radiation dose reduction compared to digital radiographs (DR). Three groups, each consisting of 19 patients, were examined with different DSCT protocols using tin prefiltration (Sn96/64/32 ref. mAs at 100 kV). Different strengths of iterative reconstruction were applied (ADMIRE 2/3/4). DSCT groups were compared to 19 matched patients examined with posterior–anterior DR. Diagnostic confidence, detectability of anatomical structures and small lung lesions were evaluated on a 4-point Likert scale (LS 1 = unacceptable, 4 = fully acceptable; a value ≥ 3 was considered acceptable). Effective dose (ED) was 31-/21-/9-fold higher in Sn96/Sn64/Sn32 compared to DR. Diagnostic confidence was sufficient in Sn96/Sn64 (LS 3.4/3.2), reduced in Sn32 (LS 2.7) and the worst in DR (LS 2.4). In DSCT, detectability of small anatomical structures was always superior to DR (p < 0.05). Mean lesion size ranged from 5.1–7 mm; detectability was acceptable in all DSCT groups (LS 3.0–3.4) and superior to DR (LS 1.9; p < 0.05). Substantial dose lowering in DSCT of the pediatric lung enables acceptable detectability of small lung lesions with a radiation dose being about 10-fold higher compared to DR. [ABSTRACT FROM AUTHOR]