Manifold Learning-based Data Sampling for Model Training

Automatic dementia screening and scoring by applying deep learning on clock-drawing tests

Dementia is one of the most common neurological syndromes in the world. Usually, diagnoses are made based on paper-and-pencil tests and scored depending on personal judgments of experts. This technique can introduce errors and has high inter-rater variability. To overcome these issues, we present an automatic assessment of the widely used paper-based clock-drawing test by means of deep neural networks. Our study includes a comparison of three modern architectures: VGG16, ResNet-152, and DenseNet-121. The dataset consisted of 1315 individuals. To deal with the limited amount of data, which also included several dementia types, we used optimization strategies for training the neural network. The outcome of our work is a standardized and digital estimation of the dementia screening result and severity level for an individual. We achieved accuracies of 96.65% for screening and up to 98.54% for scoring, overcoming the reported state-of-the-art as well as human accuracies. Due to the digital format, the paper-based test can be simply scanned by using a mobile device and then be evaluated also in areas where there is a staff shortage or where no clinical experts are available.

Automatic multi-organ segmentation in dual-energy CT (DECT) with dedicated 3D fully convolutional DECT networks

PURPOSE

Dual-energy computed tomography (DECT) has shown great potential in many clinical applications. By incorporating the information from two different energy spectra, DECT provides higher contrast and reveals more material differences of tissues compared to conventional single-energy CT (SECT). Recent research shows that automatic multi-organ segmentation of DECT data can improve DECT clinical applications. However, most segmentation methods are designed for SECT, while DECT has been significantly less pronounced in research. Therefore, a novel approach is required that is able to take full advantage of the extra information provided by DECT.

METHODS

In the scope of this work, we proposed four three-dimensional (3D) fully convolutional neural network algorithms for the automatic segmentation of DECT data. We incorporated the extra energy information differently and embedded the fusion of information in each of the network architectures.

RESULTS

Quantitative evaluation using 45 thorax/abdomen DECT datasets acquired with a clinical dual-source CT system was investigated. The segmentation of six thoracic and abdominal organs (left and right lungs, liver, spleen, and left and right kidneys) were evaluated using a fivefold cross-validation strategy. In all of the tests, we achieved the best average Dice coefficients of 98% for the right lung, 98% for the left lung, 96% for the liver, 92% for the spleen, 95% for the right kidney, 93% for the left kidney, respectively. The network architectures exploit dual-energy spectra and outperform deep learning for SECT.

CONCLUSIONS

The results of the cross-validation show that our methods are feasible and promising. Successful tests on special clinical cases reveal that our methods have high adaptability in the practical application.