Search

We demonstrate deep-learning neural network (NN)-based dynamic optical coherence tomography (DOCT), which generates high-quality logarithmic-intensity-variance (LIV) DOCT images from only four OCT frames. The NN model is trained for tumor spheroid samples using a customized loss function: the weighted mean absolute error. This loss function enables highly accurate LIV image generation. The fidelity of the generated LIV images to the ground truth LIV images generated using 32 OCT frames is examined via subjective image observation and statistical analysis of image-based metrics. Fast volumetric DOCT imaging with an acquisition time of 6.55 s/volume is demonstrated using this NN-based method.

We demonstrate a deep-learning-based scatterer density estimator (SDE) that processes local speckle patterns of optical coherence tomography (OCT) images and estimates the scatterer density behind each speckle pattern. The SDE is trained using large quantities of numerically simulated OCT images and their associated scatterer densities. The numerical simulation uses a noise model that incorporates the spatial properties of three types of noise, i.e., shot noise, relative-intensity noise, and non-optical noise. The SDE's performance was evaluated numerically and experimentally using two types of scattering phantom and in vitro tumor spheroids. The results confirmed that the SDE estimates scatterer densities accurately. The estimation accuracy improved significantly when compared with our previous deep-learning-based SDE, which was trained using numerical speckle patterns generated from a noise model that did not account for the spatial properties of noise.

A new formulation of lateral imaging process of point-scanning optical coherence tomography (OCT) and a new differential contrast method designed by using this formulation are presented. The formulation is based on a mathematical sample model called the dispersed scatterer model (DSM), in which the sample is represented as a material with a spatially slowly varying refractive index and randomly distributed scatterers embedded in the material. It is shown that the formulation represents a meaningful OCT image and speckle as two independent mathematical quantities. The new differential contrast method is based on complex signal processing of OCT images, and the physical and numerical imaging processes of this method are jointly formulated using the same theoretical strategy as in the case of OCT. The formula shows that the method provides a spatially differential image of the sample structure. This differential imaging method is validated by measuring in vivo and in vitro samples.

An organoid is a three-dimensional (3D) in vitro cell culture emulating human organs. We applied 3D dynamic optical coherence tomography (DOCT) to visualize the intratissue and intracellular activities of human induced pluripotent stem cells (hiPSCs)-derived alveolar organoids in normal and fibrosis models. 3D DOCT data were acquired with an 840-nm spectral domain optical coherence tomography with axial and lateral resolutions of 3.8 {\mu}m (in tissue) and 4.9 {\mu}m, respectively. The DOCT images were obtained by the logarithmic-intensity-variance (LIV) algorithm, which is sensitive to the signal fluctuation magnitude. The LIV images revealed cystic structures surrounded by high-LIV borders and mesh-like structures with low LIV. The former may be alveoli with a highly dynamics epithelium, while the latter may be fibroblasts. The LIV images also demonstrated the abnormal repair of the alveolar epithelium.

We demonstrate label-free drug response evaluations of human breast (MCF-7) and colon (HT-29) cancer spheroids via dynamic optical coherence tomography (OCT). The MCF-7 and HT-29 spheroids were treated with paclitaxel (PTX, or Taxol) and the active metabolite of irinotecan (SN-38), respectively. The drugs were applied using 0 (control), 0.1, 1, and 10 uM concentrations with treatment times of 1, 3, and 6 days. The samples were scanned using a repeated raster scan protocol and two dynamic OCT algorithms, logarithmic intensity variance (LIV) and late OCT correlation decay speed (OCDSl) analyses, were applied to visualize the tissue and cellular dynamics. Different drug response patterns of the two spheroid types were visualized clearly and analyzed quantitatively by LIV and OCDSl imaging. For both spheroid types, structural corruptions and reduction of LIV and OCDSl were observed. These results may indicate different mechanisms of the drug action. The results suggest that dynamic OCT can be used to highlight drug response patterns and perform anti-cancer drug testing.

Label-free metabolic imaging of non-alcoholic fatty liver disease (NAFLD) mouse liver is demonstrated ex vivo by dynamic optical coherence tomography (OCT). The NAFLD mouse is a methionine choline-deficient (MCD)-diet model, and two mice fed MCD diet for 1 and 2 weeks are involved in addition to a normal-diet mouse. The dynamic OCT is based on repeating raster scan and logarithmic intensity variance (LIV) analysis which enables volumetric metabolic imaging with a standard-speed (50,000 A-lines/s) OCT system. Metabolic domains associated with lipid droplet accumulation and inflammation are clearly visualized three-dimensionally. Particularly, the normal-diet liver exhibits highly metabolic vessel-like structures of peri-vascular hepatic zones. The 1-week MCD-diet liver shows ring-shaped highly metabolic structures formed with lipid droplets. The 2-week MCD-diet liver exhibits fragmented vessel-like structures associated with inflammation. These results imply that volumetric LIV imaging is useful for visualizing and assessing NAFLD abnormalities.

Heart rate (HR) detection from ballistocardiogram (BCG) signals is challenging because the signal morphology can vary between and within-subjects. Also, it differs from one sensor to another. Hence, it is essential to evaluate HR detection algorithms across several datasets and under different experimental setups. In this paper, we studied the potential of three HR detection algorithms across four independent BCG datasets. The three algorithms are as follows: the multiresolution analysis of the maximal overlap discrete wavelet transform (MODWT-MRA), continuous wavelet transform (CWT), and template matching (TM). The four datasets were obtained using a microbend fiber optic sensor, a fiber Bragg grating sensor, electromechanical films, and load cells, respectively. The datasets were gathered from: a) 10 patients during a polysomnography study, b) 50 subjects in a sitting position, c) 10 subjects in a sleeping position, and d) 40 subjects in a sleeping position. Overall, CWT with derivative of Gaussian provided superior results compared with the MODWT-MRA, CWT (frequency B-spline), and CWT (Shannon). That said, a BCG template was constructed from DataSet1. Then, it was used for HR detection in the other datasets. The TM method achieved satisfactory results for DataSet2 and DataSet3, but it did not detect the HR of two subjects in DataSet4. The proposed methods were implemented on a Raspberry Pi. As a result, the average time required to analyze a 30-second BCG signal was less than one second for all methods. Yet, the MODWT-MRA had the highest performance with an average time of 0.04 seconds., Comment: Preprint submitted to Biomedical Signal Processing and Control Journal

Time-domain algorithms are focused on detecting local maxima or local minima using a moving window, and therefore finding the interval between the dominant J-peaks of ballistocardiogram (BCG) signal. However, this approach has many limitations due to the nonlinear and nonstationary behavior of the BCG signal. This is because the BCG signal does not display consistent J-peaks, which can usually be the case for overnight, in-home monitoring, particularly with frail elderly. Additionally, its accuracy will be undoubtedly affected by motion artifacts. Second, frequency-domain algorithms do not provide information about interbeat intervals. Nevertheless, they can provide information about heart rate variability. This is usually done by taking the fast Fourier transform or the inverse Fourier transform of the logarithm of the estimated spectrum, i.e., cepstrum of the signal using a sliding window. Thereafter, the dominant frequency is obtained in a particular frequency range. The limit of these algorithms is that the peak in the spectrum may get wider and multiple peaks may appear, which might cause a problem in measuring the vital signs. At last, the objective of wavelet-domain algorithms is to decompose the signal into different components, hence the component which shows an agreement with the vital signs can be selected i.e., the selected component contains only information about the heart cycles or respiratory cycles, respectively. An empirical mode decomposition is an alternative approach to wavelet decomposition, and it is also a very suitable approach to cope with nonlinear and nonstationary signals such as cardiorespiratory signals. Apart from the above-mentioned algorithms, machine learning approaches have been implemented for measuring heartbeats. However, manual labeling of training data is a restricting property., Comment: Review Paper

The diabetic retinopathy is timely diagonalized through color eye fundus images by experienced ophthalmologists, in order to recognize potential retinal features and identify early-blindness cases. In this paper, it is proposed to extract deep features from the last fully-connected layer of, four different, pre-trained convolutional neural networks. These features are then feeded into a non-linear classifier to discriminate three-class diabetic cases, i.e., normal, exudates, and drusen. Averaged across 1113 color retinal images collected from six publicly available annotated datasets, the deep features approach perform better than the classical bag-of-words approach. The proposed approaches have an average accuracy between 91.23% and 92.00% with more than 13% improvement over the traditional state of art methods., Comment: Preprint submitted to Journal of Medical Imaging | SPIE (Tue, Jul 28, 2017)

The main objective of this project is to segment different breast ultrasound images to find out lesion area by discarding the low contrast regions as well as the inherent speckle noise. The proposed method consists of three stages (removing noise, segmentation, classification) in order to extract the correct lesion. We used normalized cuts approach to segment ultrasound images into regions of interest where we can possibly finds the lesion, and then K-means classifier is applied to decide finally the location of the lesion. For every original image, an annotated ground-truth image is given to perform comparison with the obtained experimental results, providing accurate evaluation measures., Comment: Course Work in 'Medical Image Analysis' Module, VIBOT 2013

In literature, several approaches are trying to make the UAVs fly autonomously i.e., by extracting perspective cues such as straight lines. However, it is only available in well-defined human made environments, in addition to many other cues which require enough texture information. Our main target is to detect and avoid frontal obstacles from a monocular camera using a quad rotor Ar.Drone 2 by exploiting optical flow as a motion parallax, the drone is permitted to fly at a speed of 1 m/s and an altitude ranging from 1 to 4 meters above the ground level. In general, detecting and avoiding frontal obstacle is a quite challenging problem because optical flow has some limitation which should be taken into account i.e. lighting conditions and aperture problem.

Diabetic retinopathy (DR) and age related macular degeneration (ARMD) are among the major causes of visual impairment worldwide. DR is mainly characterized by red spots, namely microaneurysms and bright lesions, specifically exudates whereas ARMD is mainly identified by tiny yellow or white deposits called drusen. Since exudates might be the only manifestation of the early diabetic retinopathy, there is an increase demand for automatic retinopathy diagnosis. Exudates and drusen may share similar appearances, thus discriminating between them is of interest to enhance screening performance. In this research, we investigative the role of bag of words approach in the automatic diagnosis of retinopathy diabetes. We proposed to use a single based and multiple based methods for the construction of the visual dictionary by combining the histogram of word occurrences from each dictionary and building a single histogram. The introduced approach is evaluated for automatic diagnosis of normal and abnormal color fundus images with bright lesions. This approach has been implemented on 430 fundus images, including six publicly available datasets, in addition to one local dataset. The mean accuracies reported are 97.2% and 99.77% for single based and multiple based dictionaries respectively.