Wyniki wyszukiwania - Biblioteka Nauki

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

Znaleziono wyników: 3

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: medical image processing

Sortuj według:

Ogranicz wyniki do:

Quantification of the ilio-femoral arteries tortuosity and data cluster modelling for preoperative examination

100%

Markiewicz T. , Dziekiewicz M.

tom Vol. 42, no. 4

1123--1136

Detecting spatial tortuosity and atherosclerotic changes of the ilio-femoral arteries are crucial for planning endovascular access. The aim of this study was to find a reliable quantification procedure of arterial lumen and tortuosity to qualify patients for a suitable endovascular procedure. We conducted computed tomographic angiography in 76 patients. All ilio-femoral segments of the arterial tree were visualized using Osirix Dicom Viewer software to help qualify the patients to one of two groups: with possible or non-recommended vascular access. The same tomograms were then analyzed with image processing algorithms to perform ilio-femoral artery segmentation and quantification. We chose a set of arterial tortuosity and lumen measuring methods, such as the modified Gustafson-Kessel clustering algorithm and Support Vector Machine classifier, to automatically classify arterial-tree regions. The two 2D feature spaces were selected with the modified Gustafson-Kessel clusterization to create a combined model to assign around 2/3 cases to the access groups with high specificity (more than 88%) whereas the remaining patients were selected for re-evaluation. We concluded that the novel modification of the Gustafson-Kessel clustering algorithm is more suitable to the highly inseparable data than commonly used approaches. To identify ilio-femoral access limitations, we recommend more complex decision model. This study confirmed high usability of our chosen methodology in the quantitative examination of arteries for endovascular access planning.

Quantitative impact of label noise on the quality of segmentation of brain tumors on MRI scans

80%

Marcinkiewicz M. , Mrukwa G.

2019

tom Vol. 18

61--65

Over the last few years, deep learning has proven to be a great solution to many problems, such as image or text classification. Recently, deep learning-based solutions have outperformed humans on selected benchmark datasets, yielding a promising future for scientific and real-world applications. Training of deep learning models requires vast amounts of high quality data to achieve such supreme performance. In real-world scenarios, obtaining a large, coherent, and properly labeled dataset is a challenging task. This is especially true in medical applications, where high-quality data and annotations are scarce and the number of expert annotators is limited. In this paper, we investigate the impact of corrupted ground-truth masks on the performance of a neural network for a brain tumor segmentation task. Our findings suggest that a) the performance degrades about 8% less than it could be expected from simulations, b) a neural network learns the simulated biases of annotators, c) biases can be partially mitigated by using an inversely-biased dice loss function.

Machine-learning at the service of plastic surgery: a case study evaluating facial attractiveness and emotions using R language

61%

Štěpánek L. , Měšťák J. , Kasal P.

2019

tom Vol. 18

107--112

Since the plastic surgery should consider that facial impression is always dependent on current facial emotion, it came to be verified how precise classification of facial images into sets of defined facial emotions is.