Output list
Conference proceeding
Published 2018
Image Analysis for Moving Organ, Breast, and Thoracic Images, 11040, 191 - 201
International Workshop on Reconstruction and Analysis of Moving Body Organs
Deep learning using convolutional neural networks (ConvNets) achieves high accuracy across many computer vision tasks, with the ability to learn multi-scale features and generalize across a variety of input data. In this work, we propose a deep learning framework that utilizes a coarse-to-fine cascade of 3D ConvNet models for segmentation of lung structures obtained from computed tomographic (CT) images. Deep learning requires a large number of training datasets, which may be challenging in medical imaging, especially for rare diseases. In the present study, transfer learning is utilized for lung segmentation of CT scans in large animal models of the acute respiratory distress syndrome (ARDS) using only 13 subjects. The method was quantitatively evaluated on a human dataset, consisting of 395 3D CT scans from 153 subjects, and an animal dataset consisting of 148 3D CT images from 5 porcine subjects. The human dataset achieved an average Jacaard index of 0.99, and an average symmetric surface distance (ASSD) of 0.29 mm. The animal dataset had an average Jacaard index of 0.94, and an ASSD of 0.99 mm.
Conference proceeding
Published 1999
Proceedings of the First Joint BMES/EMBS Conference. 1999 IEEE Engineering in Medicine and Biology 21st Annual Conference and the 1999 Annual Fall Meeting of the Biomedical Engineering Society (Cat. N, 1, 337 vol.1 - 337
The authors have advanced an algorithm that can use measurements of pressure and flow entering a standard flexible patient tubing circuit during clinical ventilation and predict pressures and flows actually delivered to a patient. For volume ventilation and step, ramp, or sinusoidal waveforms, predictions are within 5% of truth for most of the inspiratory cycle. From these data, the authors can then accurately track respiratory dynamic resistance and elastance.
Conference proceeding
Published 1991
Proceedings of the 1991 IEEE Seventeenth Annual Northeast Bioengineering Conference, 73 - 74
Frequency domain and offline and online time domain techniques are compared to estimate respiratory mechanical properties using standard ventilator waveforms at low frequencies. Both methods are shown to yield consistent parameter estimates for a simple series RC model and pure sine wave input. However, they could not accurately predict frequency dependence of respiratory resistance from 0 to 1 Hz when using a step flow waveform. This behavior is traced to signal-to-noise limitations and nonlinearities.< >