Toward real-time polyp detection using fully CNNs for 2D Gaussian shapes prediction.

To decrease colon polyp miss-rate during colonoscopy, a real-time detection system with high accuracy is needed. Recently, there have been many efforts to develop models for real-time polyp detection, but work is still required to develop real-time detection algorithms with reliable results. We use single-shot feed-forward fully convolutional neural networks (F-CNN) to develop an accurate real-time polyp detection system. F-CNNs are usually trained on binary masks for object segmentation. We propose the use of 2D Gaussian masks instead of binary masks to enable these models to detect different types of polyps more effectively and efficiently and reduce the number of false positives. The experimental results showed that the proposed 2D Gaussian masks are efficient for detection of flat and small polyps with unclear boundaries between background and polyp parts. The masks make a better training effect to discriminate polyps from the polyp-like false positives. The proposed method achieved state-of-the-art results on two polyp datasets. On the ETIS-LARIB dataset we achieved 86.54% recall, 86.12% precision, and 86.33% F1-score, and on the CVC-ColonDB we achieved 91% recall, 88.35% precision, and F1-score 89.65%.

Improving Automatic Polyp Detection Using CNN by Exploiting Temporal Dependency in Colonoscopy Video.

Automatic polyp detection has been shown to be difficult due to various polyp-like structures in the colon and high interclass variations in polyp size, color, shape, and texture. An efficient method should not only have a high correct detection rate (high sensitivity) but also a low false detection rate (high precision and specificity). The state-of-the-art detection methods include convolutional neural networks (CNN). However, CNNs have shown to be vulnerable to small perturbations and noise; they sometimes miss the same polyp appearing in neighboring frames and produce a high number of false positives. We aim to tackle this problem and improve the overall performance of the CNN-based object detectors for polyp detection in colonoscopy videos. Our method consists of two stages: a region of interest (RoI) proposal by CNN-based object detector networks and a false positive (FP) reduction unit. The FP reduction unit exploits the temporal dependencies among image frames in video by integrating the bidirectional temporal information obtained by RoIs in a set of consecutive frames. This information is used to make the final decision. The experimental results show that the bidirectional temporal information has been helpful in estimating polyp positions and accurately predict the FPs. This provides an overall performance improvement in terms of sensitivity, precision, and specificity compared to conventional false positive learning method, and thus achieves the state-of-the-art results on the CVC-ClinicVideoDB video data set.

Two-Wired Active Spring-Loaded Dry Electrodes for EEG Measurements.

Dry contact electrode-based EEG acquisition is one of the easiest ways to obtain neural information from the human brain, providing many advantages such as rapid installation, and enhanced wearability. However, high contact impedance due to insufficient electrical coupling at the electrode-scalp interface still remains a critical issue. In this paper, a two-wired active dry electrode system is proposed by combining finger-shaped spring-loaded probes and active buffer circuits. The shrinkable probes and bootstrap topology-based buffer circuitry provide reliable electrical coupling with an uneven and hairy scalp and effective input impedance conversion along with low input capacitance. Through analysis of the equivalent circuit model, the proposed electrode was carefully designed by employing off-the-shelf discrete components and a low-noise zero-drift amplifier. Several electrical evaluations such as noise spectral density measurements and input capacitance estimation were performed together with simple experiments for alpha rhythm detection. The experimental results showed that the proposed electrode is capable of clear detection for the alpha rhythm activation, with excellent electrical characteristics such as low-noise of 1.131 µVRMS and 32.3% reduction of input capacitance.

Dry Electrode-Based Fully Isolated EEG/fNIRS Hybrid Brain-Monitoring System.

A portable hybrid brain monitoring system is proposed to perform simultaneous 16-channel electroencephalogram (EEG) and 8-channel functional near-infrared spectroscopy (fNIRS) measurements. Architecture-optimized analog frontend integrated circuits (Texas Instruments ADS1299 and ADS8688A) were used to simultaneously achieve 24-bit EEG resolution and reliable latency-less (<0.85 µs) bio-optical measurements. Suppression of the noise and crosstalk generated by the digital circuit components and flashing NIR light sources was maximized through linear regulator-based fully isolated circuit design. Gel-less EEG measurements were enabled by using spring-loaded dry electrodes. Several evaluations were carried out by conducting an EEG phantom test and an arterial occlusion experiment. An alpha rhythm detection test (eye-closing task) and a mental arithmetic experiment (cumulative subtraction task) were conducted to determine whether the system is applicable to human subject studies. The evaluation results show that the proposed system is sufficiently capable of detecting microvoltage EEG signals and hemodynamic responses. The results of the studies on human subjects enabled us to verify that the proposed system is able to detect task-related EEG spectral features such as eye-closed event-related synchronization and mental-arithmetic event-related desynchronization in the alpha and beta rhythm ranges. An analysis of the fNIRS measurements with an arithmetic operation task also revealed a decreasing trend in oxyhemoglobin concentration.

Automatic polyp frame screening using patch based combined feature and dictionary learning.

Polyps in the colon can potentially become malignant cancer tissues where early detection and removal lead to high survival rate. Certain types of polyps can be difficult to detect even for highly trained physicians. Inspired by aforementioned problem our study aims to improve the human detection performance by developing an automatic polyp screening framework as a decision support tool. We use a small image patch based combined feature method. Features include shape and color information and are extracted using histogram of oriented gradient and hue histogram methods. Dictionary learning based training is used to learn features and final feature vector is formed using sparse coding. For classification, we use patch image classification based on linear support vector machine and whole image thresholding. The proposed framework is evaluated using three public polyp databases. Our experimental results show that the proposed scheme successfully classified polyps and normal images with over 95% of classification accuracy, sensitivity, specificity and precision. In addition, we compare performance of the proposed scheme with conventional feature based methods and the convolutional neural network (CNN) based deep learning approach which is the state of the art technique in many image classification applications.

Simple adaptive sparse representation based classification schemes for EEG based brain-computer interface applications.

One of the main problems related to electroencephalogram (EEG) based brain-computer interface (BCI) systems is the non-stationarity of the underlying EEG signals. This results in the deterioration of the classification performance during experimental sessions. Therefore, adaptive classification techniques are required for EEG based BCI applications. In this paper, we propose simple adaptive sparse representation based classification (SRC) schemes. Supervised and unsupervised dictionary update techniques for new test data and a dictionary modification method by using the incoherence measure of the training data are investigated. The proposed methods are very simple and additional computation for the re-training of the classifier is not needed. The proposed adaptive SRC schemes are evaluated using two BCI experimental datasets. The proposed methods are assessed by comparing classification results with the conventional SRC and other adaptive classification methods. On the basis of the results, we find that the proposed adaptive schemes show relatively improved classification accuracy as compared to conventional methods without requiring additional computation.

Sparse representation-based classification scheme for motor imagery-based brain-computer interface systems.

Motor imagery (MI)-based brain-computer interface systems (BCIs) normally use a powerful spatial filtering and classification method to maximize their performance. The common spatial pattern (CSP) algorithm is a widely used spatial filtering method for MI-based BCIs. In this work, we propose a new sparse representation-based classification (SRC) scheme for MI-based BCI applications. Sensorimotor rhythms are extracted from electroencephalograms and used for classification. The proposed SRC method utilizes the frequency band power and CSP algorithm to extract features for classification. We analyzed the performance of the new method using experimental datasets. The results showed that the SRC scheme provides highly accurate classification results, which were better than those obtained using the well-known linear discriminant analysis classification method. The enhancement of the proposed method in terms of the classification accuracy was verified using cross-validation and a statistical paired t-test (p < 0.001).

Engineering the Japanese encephalitis virus RNA genome for the expression of foreign genes of various sizes: implications for packaging capacity and RNA replication efficiency.

Using the RNA replication machinery of Japanese encephalitis virus (JEV), the authors have established and characterized three strategies for the expression of foreign genes. Initially, approximately 11 kb genomic RNA was engineered to express heterologous genes of various sizes by preferentially inserting a new cistron at the beginning of the 3' nontranslated variable region. RNA transfection yielded recombinant viruses that initiated foreign gene expression after infecting permissive cells. JEV was capable of packaging recombinant genomes as large as approximately 15 kb. However, larger genome size was inversely correlated with RNA replication efficiency and cytopathogenicity, with no significant change in infectivity. Second, a variety of self-replicating propagation-deficient viral replicons were constructed by introducing one to three in-frame deletions into the ectodomains of all the structural proteins of JEV. These replicons displayed a spectrum of RNA replication efficiency upon transfection, suggesting that remnant transmembrane domains play a suppressive role in this process. Third, the authors generated a panel of stable packaging cell lines (PCLs) providing all three JEV structural proteins in trans. These PCLs efficiently packaged viral replicon RNAs into single-round infectious viral replicon particles. These JEV-based virus/vector systems may provide useful tools for a variety of biological applications, including foreign gene expression, antiviral compound screening, and genetic immunization.