Universitätspublikationen
Refine
Document Type
- Article (4)
Language
- English (4)
Has Fulltext
- yes (4) (remove)
Is part of the Bibliography
- no (4)
Keywords
Institute
- Medizin (3)
- E-Finance Lab e.V. (1)
- Informatik und Mathematik (1)
Background: Computed tomography (CT) low-dose (LD) imaging is used to lower radiation exposure, especially in vascular imaging; in current literature, this is mostly on latest generation high-end CT systems.
Purpose: To evaluate the effects of reduced tube current on objective and subjective image quality of a 15-year-old 16-slice CT system for pulmonary angiography (CTPA).
Material and Methods: CTPA scans from 60 prospectively randomized patients (28 men, 32 women) were examined in this study on a 15-year-old 16-slice CT scanner system. Standard CT (SD) settings were 100 kV and 150 mAs, LD settings were 100 kV and 50 mAs. Attenuation of the pulmonary trunk, various anatomic landmarks, and image noise were quantitatively measured; contrast-to-noise ratios (CNR) and signal-to-noise ratios (SNR) were calculated. Three independent blinded radiologists subjectively rated each image series using a 5-point grading scale.
Results: CT dose index (CTDI) in the LD series was 66.46% lower compared to the SD settings (2.49 ± 0.55 mGy versus 7.42 ± 1.17 mGy). Attenuation of the pulmonary trunk showed similar results for both series (SD 409.55 ± 91.04 HU; LD 380.43 HU ± 93.11 HU; P = 0.768). Subjective image analysis showed no significant differences between SD and LD settings regarding the suitability for detection of central and peripheral PE (central SD/LD, 4.88; intra-class correlation coefficients [ICC], 0.894/4.83; ICC, 0.745; peripheral SD/LD, 4.70; ICC, 0.943/4.57; ICC, 0.919; all P > 0.4).
Conclusion: The LD protocol, on a 15-year-old CT scanner system without current high-end hardware or post-processing tools, led to a dose reduction of approximately 67% with similar subjective image quality and delineation of central and peripheral pulmonary arteries.
COVID-19 HAS AGAIN TIGHTENED ITS GRIP AROUND THE WORLD AND THE HEALTH SYSTEM. THIS ARTICLE GIVES AN INTRODUCTION TO EXPLAINABLE INTERACTIVE MACHINE LEARNING AND PROVIDES INSIGHTS ON HOW THIS METHOD MAY NOT ONLY HELP IN ENGINEERING MORE POWERFUL AI SYSTEMS, BUT ALSO HOW IT MAY HELP TO EASE THE BURDEN OF VIRAL STRAINS ON THE HEALTHCARE SYSTEM.
DNA methylation-based prediction of response to immune checkpoint inhibition in metastatic melanoma
(2021)
Background: Therapies based on targeting immune checkpoints have revolutionized the treatment of metastatic melanoma in recent years. Still, biomarkers predicting long-term therapy responses are lacking. Methods: A novel approach of reference-free deconvolution of large-scale DNA methylation data enabled us to develop a machine learning classifier based on CpG sites, specific for latent methylation components (LMC), that allowed for patient allocation to prognostic clusters. DNA methylation data were processed using reference-free analyses (MeDeCom) and reference-based computational tumor deconvolution (MethylCIBERSORT, LUMP). Results: We provide evidence that DNA methylation signatures of tumor tissue from cutaneous metastases are predictive for therapy response to immune checkpoint inhibition in patients with stage IV metastatic melanoma. Conclusions: These results demonstrate that LMC-based segregation of large-scale DNA methylation data is a promising tool for classifier development and treatment response estimation in cancer patients under targeted immunotherapy.
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.