Refine
Year of publication
- 2020 (5) (remove)
Document Type
- Article (5)
Has Fulltext
- yes (5)
Is part of the Bibliography
- no (5)
Keywords
Institute
- Medizin (5)
- Informatik und Mathematik (1)
Background: Myocardial perfusion with cardiovascular magnetic resonance (CMR) imaging is an established diagnostic test for evaluation of myocardial ischaemia. For quantification purposes, the 16 segment American Heart Association (AHA) model poses limitations in terms of extracting relevant information on the extent/severity of ischaemia as perfusion deficits will not always fall within an individual segment, which reduces its diagnostic value, and makes an accurate assessment of outcome data or a result comparison across various studies difficult. We hypothesised that division of the myocardial segments into epi- and endocardial layers and a further circumferential subdivision, resulting in a total of 96 segments, would improve the accuracy of detecting myocardial hypoperfusion. Higher (sub-)subsegmental recording of perfusion abnormalities, which are defined relatively to the normal reference using the subsegment with the highest value, may improve the spatial encoding of myocardial blood flow, based on a single stress perfusion acquisition. Objective: A proof of concept comparison study of subsegmentation approaches based on transmural segments (16 AHA and 48 segments) vs. subdivision into epi- and endocardial (32) subsegments vs. further circumferential subdivision into 96 (sub-)subsegments for diagnostic accuracy against invasively defined obstructive coronary artery disease (CAD). Methods: Thirty patients with obstructive CAD and 20 healthy controls underwent perfusion stress CMR imaging at 3 T during maximal adenosine vasodilation and a dual bolus injection of 0.1mmol/kg gadobutrol. Using Fermi deconvolution for blood flow estimation, (sub-)subsegmental values were expressed relative to the (sub)subsegment with the highest flow. In addition, endo−/epicardial flow ratios were calculated based on 32 and 96 (sub-)subsegments. A receiver operating characteristics (ROC) curve analysis was performed to compare the diagnostic performance of discrimination between patients with CAD and healthy controls. Observer reproducibility was assessed using Bland-Altman approaches. Results: Subdivision into more and smaller segments revealed greater accuracy for #32, #48 and # 96 compared to the standard #16 approach (area under the curve (AUC): 0.937, 0.973 and 0.993 vs 0.820, p<0.05). The #96-based endo−/epicardial ratio was superior to the #32 endo−/epicardial ratio (AUC 0.979, vs. 0.932, p<0.05). Measurements for the #16 model showed marginally better reproducibility compared to #32, #48 and #96 (mean difference± standard deviation: 2.0±3.6 vs. 2.3±4.0 vs 2.5±4.4 vs. 4.1±5.6). Conclusions: Subsegmentation of the myocardium improves diagnostic accuracy and facilitates an objective cutoff-based description of hypoperfusion, and facilitates an objective description of hypoperfusion, including the extent and severity of myocardial ischaemia. Quantification based on a single (stress-only) pass reduces the overall amount of gadolinium contrast agent required and the length of the overall diagnostic study.
Background: Bone age (BA) assessment performed by artificial intelligence (AI) is of growing interest due to improved accuracy, precision and time efficiency in daily routine. The aim of this study was to investigate the accuracy and efficiency of a novel AI software version for automated BA assessment in comparison to the Greulich-Pyle method.
Methods: Radiographs of 514 patients were analysed in this retrospective study. Total BA was assessed independently by three blinded radiologists applying the GP method and by the AI software. Overall and gender-specific BA assessment results, as well as reading times of both approaches, were compared, while the reference BA was defined by two blinded experienced paediatric radiologists in consensus by application of the Greulich-Pyle method.
Results: Mean absolute deviation (MAD) and root mean square deviation (RSMD) were significantly lower between AI-derived BA and reference BA (MAD 0.34 years, RSMD 0.38 years) than between reader-calculated BA and reference BA (MAD 0.79 years, RSMD 0.89 years; p < 0.001). The correlation between AI-derived BA and reference BA (r = 0.99) was significantly higher than between reader-calculated BA and reference BA (r = 0.90; p < 0.001). No statistical difference was found in reader agreement and correlation analyses regarding gender (p = 0.241). Mean reading times were reduced by 87% using the AI system.
Conclusions: A novel AI software enabled highly accurate automated BA assessment. It may improve efficiency in clinical routine by reducing reading times without compromising the accuracy compared with the Greulich-Pyle method.
Objectives: To analyze the performance of radiological assessment categories and quantitative computational analysis of apparent diffusion coefficient (ADC) maps using variant machine learning algorithms to differentiate clinically significant versus insignificant prostate cancer (PCa). Methods: Retrospectively, 73 patients were included in the study. The patients (mean age, 66.3 ± 7.6 years) were examined with multiparametric MRI (mpMRI) prior to radical prostatectomy (n = 33) or targeted biopsy (n = 40). The index lesion was annotated in MRI ADC and the equivalent histologic slides according to the highest Gleason Grade Group (GrG). Volumes of interest (VOIs) were determined for each lesion and normal-appearing peripheral zone. VOIs were processed by radiomic analysis. For the classification of lesions according to their clinical significance (GrG ≥ 3), principal component (PC) analysis, univariate analysis (UA) with consecutive support vector machines, neural networks, and random forest analysis were performed. Results: PC analysis discriminated between benign and malignant prostate tissue. PC evaluation yielded no stratification of PCa lesions according to their clinical significance, but UA revealed differences in clinical assessment categories and radiomic features. We trained three classification models with fifteen feature subsets. We identified a subset of shape features which improved the diagnostic accuracy of the clinical assessment categories (maximum increase in diagnostic accuracy ΔAUC = + 0.05, p < 0.001) while also identifying combinations of features and models which reduced overall accuracy. Conclusions: The impact of radiomic features to differentiate PCa lesions according to their clinical significance remains controversial. It depends on feature selection and the employed machine learning algorithms. It can result in improvement or reduction of diagnostic performance.
hintergrund: Männer in Deutschland sterben früher als Frauen und nehmen weniger häufig Krebsvorsorgeuntersuchungen wahr.
Fragestellung: Ziel war die prospektive Evaluation einer „Movember-Gesundheitsinitiative“ am Universitätsklinikum Frankfurt (UKF) im November 2019.
Methoden: Im Rahmen der „Movember-Gesundheitsinitiative“ wurde allen männlichen Mitarbeitern des UKF ab dem 45. Lebensjahr und bei erstgradiger familiärer Vorbelastung eines Prostatakarzinoms ab dem 40. Lebensjahr im November 2019 gemäß S3-Leitlinien der Deutschen Gesellschaft für Urologie (DGU) eine Prostatakarzinom-Vorsorgeuntersuchung angeboten.
Ergebnisse: Insgesamt nahmen 14,4 % der Mitarbeiter teil. Eine familiäre Vorbelastung gaben insgesamt 14,0 % Teilnehmer an. Das mediane Alter betrug 54 Jahre. Der mediane PSA(prostataspezifisches Antigen)-Wert lag bei 0,9 ng/ml, der mediane PSA-Quotient bei 30 %. Bei 5 % (n = 6) zeigte sich ein suspekter Tastbefund in der DRU (digital-rektale Untersuchung). Nach Altersstratifizierung (≤ 50 vs. > 50 Lebensjahre) zeigten sich signifikante Unterschiede im medianen PSA-Wert (0,7 ng/ml vs. 1,0 ng/ml, p < 0,01) und der bereits zuvor durchgeführten urologischen Vorsorge (12,1 vs. 42,0 %, p < 0,01). Vier Teilnehmer (3,3 %) zeigten erhöhte Gesamt-PSA-Werte. Bei 32,2 % der Teilnehmer zeigte sich mindestens ein kontrollbedürftiger Befund. Insgesamt wurden 6 Prostatabiopsien durchgeführt. Hierbei zeigte sich in einem Fall ein intermediate-risk Prostatakarzinom (Gleason 3 + 4, pT3a, pPn1, pNx, R0).
Schlussfolgerungung: Im Rahmen der UKF-Movember-Gesundheitsinitiative 2019 konnten durch ein Vorsorgeangebot 121 Männer für eine Prostatakrebs-Vorsorge inklusive PSA-Testung gewonnen werden. Auffällige/kontrollbedürftige Befunde zeigten sich bei 32,2 %. Bei einem Mitarbeiter wurde ein therapiebedürftiges Prostatakarzinom entdeckt und therapiert.
The aim of the study was to obtain volumetric data of the components of the inner ear using three-dimensional reconstruction of high-resolution cone-beam computed tomography (CBCT) images. Two hundred three CBCT image series of the temporal bone from 118 anatomically normal patients (55 women and 63 men; mean age: 49.4 ± 20.4 years) with different suspected disorders were included in this study. Normative volumetric measurements of the inner ear, the cochlea, the semicircular canals (SSC), and the vestibule were determined using a semi-automated reconstruction method of the Workstation. Volumetric measurements were successfully completed in all 118 patients. Mean inner ear, cochlear, and vestibule volumes were statistically significantly larger in males than in females on both sides (p < 0.001). Regarding the semicircular canals, no statistically significant (p = 0.053) volume difference was found. The difference between the volumes on both sides was not significant. No correlation between the patient’s age and the volume of the compartments was seen (p > 0.05). There was no significant difference between mean bony inner ear volumes when the clinical diagnoses were compared (p > 0.05 for all clinical diagnoses and volumes). Our study concluded that three-dimensional reconstruction and assessment of the volumetric measurements of the inner ear can be obtained using high-resolution CBCT imaging.