Dielectrophoresis is a technology that uses the electrical properties of cells to control the movement of cells in a non-contact manner. It is important to observe cell movement in order to analyze cell characteristics using DEP technology. We developed an algorithm that can track the movement of hundreds of unlabeled cells by DEP force. The proposed algorithm consists of a cell detection step using a deep learning detection model and a cell tracking step based on a multiscale region of interest. Cell detection and tracking accuracy using Recall, precision, f-measure, and MOTA on a timelapse microscope image dataset has an accuracy of about 97% or more. In conclusion, by developing an automated tool that can perform imaging-based DEP cell analysis, cell tracking algorithms that can track hundreds of cells simultaneously can reduce cell analysis time and labor.
Recently, as the usage of electronic devices increase, modern people suffer from eye diseases. We analyzed goblet cells of wide-field fluorescence microscopy with a deep learning. In this study, we propose to real-time analysis using knowledge distillation using proposed loss function and optimized network. In the proposed method, residual based UNet was used as the teacher network to distill knowledge into lightweight E-Net. We train the student network using pixelwise loss and . The proposed method showed 4% improvements in dice-score compared to the lightweight E-Net, and the processing time was decreased to 68% compared to the case where only the teacher network was performed.
Nystagmus is a periodic, involuntary movement of eyes examined for diagnosis of various vestibular diseases such as benign paroxysmal positional vertigo, the most frequent vestibular disorder. In recent years, videonystagmography has been widely used in the examination of nystagmus due to its non-invasive feature. However, identifying and classifying nystagmus still requires professional knowledge and training. To this end, a pupil tracking algorithm was proposed in this paper using convolutional neural networks. U-Net was selected for pupil segmentation, and we constructed the ground truth of a new dataset for the training procedure. An additional tracking algorithm was designed to prevent false outputs of the U-Net model. Results show that the proposed pupil tracking algorithm scored higher performance than conventional methods.
Arrhythmia is the heartbeat losing its regularity or deviating from its average number. Among the types of arrhythmia is atrial fibrillation (AF) and atrial flutter (AFL), which are considered risk factors for development due to high morbidity and mortality. The early detection of AF/AFL is essential because their effects on the heart or complications appear after a considerable time. Electrocardiography (ECG) is a widely used screening method in primary care because of its low cost and convenience. ECG records the heart's electrical activity for a period of time via electrodes attached to the body. Owing to the development of computing power and interest in big data, attempts at deep learning (DL) have increased. The transformer was proposed by Google in 2017 and has achieved state-of-the-art performance in natural language processing. Various transformer-based models have been applied to various tasks beyond natural language processing and have shown promising prospects. However, there have been few cases of vision transformer (ViT) applications in ECG domain. It was difficult to determine whether ViT had sufficient influence in ECG domain. This study determined whether our extensive ECG dataset could make an AF/AFL diagnosis. We also confirmed whether the recently proposed ViT has AF/AFL diagnostic power.
Traditional methods of wound diagnosis have been diagnosed and prescribed by the naked eye of an expert. If the wound segmentation algorithm is applied to the wound diagnosis, the area of wound can be quantitated and used as an auxiliary means of treatment. Even with dramatic development of Deep learning technology in recent years, However, a lack of datasets generally occurs overfitting problem of deep learning model, which leads to poor performance for external datasets. Therefore, we trained the wound segmentation model by adding a new wound dataset in addition to the existing Open dataset, the Diabetic Foot Ulcer Challenge Dataset. Machine learning based methods are used when producing new dataset, ground truth images. Thus, in addition to the manual methods, Gradient Vector Flow machine learning techniques is used for ground-truth image production to reduce the time consumed in vain. The wound segmentation model used in this study is a U-net with residual block combined with cross entropy loss and Dice loss. As a result of the experiment, the wound segmentation accuracy was about 90% for Dice coefficient
KEYWORDS: Melanoma, Deep learning, Skin cancer, Medicine, Medical research, Education and training, Dermatology, Data modeling, Medical imaging, Image analysis
In Asians, melanoma appears as pigmented lesions on the hands and feet, and is often diagnosed as acral malignant melanoma (ALM) in the late stage with a very poor prognosis. Among diverse clinical characteristics of melanoma, the presence of basement membrane involvement is one of the most important prognostic factors. However, there have been few studies reporting artificial intelligence for prediction of basement membrane involvement in ALMs beyond its diagnosis. Therefore, in this study, we present a deep learning model that predicts the basement membrane involvement of ALMs from dermoscopy images.
Fluorescence lifetime imaging microscopy (FLIM) is a microscopic imaging technique to present an image of fluorophore lifetimes. It circumvents the problems of typical imaging methods such as intensity attenuation from depth since a lifetime is independent of the excitation intensity or fluorophore concentration. The lifetime is estimated from the time sequence of photon counts observed with signal-dependent noise, which has a Poisson distribution. Conventional methods usually estimate single or biexponential decay parameters. However, a lifetime component has a distribution or width, because the lifetime depends on macromolecular conformation or inhomogeneity. We present a novel algorithm based on a sparse representation which can estimate the distribution of lifetime. We verify the enhanced performance through simulations and experiments.
Color transforms are important methods in the analysis and processing of images. Image color transform and its inverse transform should be reversible for lossless image processing applications. However, color conversions are not reversible due to finite precision of the conversion coefficients. To overcome this limitation, reversible color transforms have been developed. Color integer transform requires multiplications of coefficients, which are implemented with shift and add operations in most cases. We propose to use canonical signed digit (CSD) representation of reversible color transform coefficients and exploitation of their common subexpressions to reduce the complexity of the hardware implementation significantly. We demonstrate roughly 50% reduction in computation with the proposed method.
KEYWORDS: Cameras, Databases, Chromium, Image processing, Digital cameras, Human vision and color perception, Image enhancement, Image quality, Data modeling, Digital imaging
We present a color correction algorithm for histogram equalized images captured by a digital camera. Current color
correction methods are based on human color perception of luminance and hue. However, these techniques do not
consider nonlinear camera characteristics, therefore the resulting color images show color distortions where brightness
modification is severe. We propose a new effective color correction method that depends on the camera brightness and
color curve. It utilizes the relationship of luminance and color variation of a camera used for image capture. We can
predict chrominance variation after luminance change by tracing the brightness-chrominance curve of the camera model.
Therefore the resulting image shows color that would have been obtained at different exposure using the same camera.
We verify that the processed images have natural color and that they are similar to images taken at different exposure
conditions. Moreover it is possible to apply the proposed method to software bracketing; we can change the exposure
condition of an image at post processing stage. All test results demonstrate that our method is accurate and useful in the
enhancement of a color of digital images.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.