scholarly journals Real-time deep learning-based colorectal polyp localization on clinical video footage achievable with a wide array of hardware configurations

2021 ◽  
Vol 09 (05) ◽  
pp. E741-E748
Author(s):  
Jeremi Podlasek ◽  
Mateusz Heesch ◽  
Robert Podlasek ◽  
Wojciech Kilisiński ◽  
Rafał Filip
Keyword(s):  
Neural Network ◽  
Real Time ◽  
False Positive ◽  
Network Architecture ◽  
Detection System ◽  
False Positive Rate ◽  
Computer Assisted ◽  
Processing Unit ◽  
Polyp Detection ◽  
Public Datasets

Abstract Background and study aims Several computer-assisted polyp detection systems have been proposed, but they have various limitations, from utilizing outdated neural network architectures to a requirement for multi-graphics processing unit (GPU) processing, to validating on small or non-robust datasets. To address these problems, we developed a system based on a state-of-the-art convolutional neural network architecture able to detect polyps in real time on a single GPU and tested on both public datasets and full clinical examination recordings. Methods The study comprised 165 colonoscopy procedure recordings and 2678 still photos gathered retrospectively. The system was trained on 81,962 polyp frames in total and then tested on footage from 42 colonoscopies and CVC-ClinicDB, CVC-ColonDB, Hyper-Kvasir, and ETIS-Larib public datasets. Clinical videos were evaluated for polyp detection and false-positive rates whereas the public datasets were assessed for F1 score. The system was tested for runtime performance on a wide array of hardware. Results The performance on public datasets varied from an F1 score of 0.727 to 0.942. On full examination videos, it detected 94 % of the polyps found by the endoscopist with a 3 % false-positive rate and identified additional polyps that were missed during initial video assessment. The system’s runtime fits within the real-time constraints on all but one of the hardware configurations. Conclusions We have created a polyp detection system with a post-processing pipeline that works in real time on a wide array of hardware. The system does not require extensive computational power, which could help broaden the adaptation of new commercially available systems.

Real-time Detection of Aortic Valve in Echocardiography using Convolutional Neural Networks

2020 ◽  
Vol 16 (5) ◽  
pp. 584-591 ◽  
Author(s):  
Muhammad Hanif Ahmad Nizar ◽  
Chow Khuen Chan ◽  
Azira Khalil ◽  
Ahmad Khairuddin Mohamed Yusof ◽  
Khin Wee Lai
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Heart Disease ◽  
Aortic Valve ◽  
Real Time ◽  
Convolutional Neural Networks ◽  
Valvular Heart Disease ◽  
Detection System ◽  
Processing Unit ◽  
Real Time Detection

Background: Valvular heart disease is a serious disease leading to mortality and increasing medical care cost. The aortic valve is the most common valve affected by this disease. Doctors rely on echocardiogram for diagnosing and evaluating valvular heart disease. However, the images from echocardiogram are poor in comparison to Computerized Tomography and Magnetic Resonance Imaging scan. This study proposes the development of Convolutional Neural Networks (CNN) that can function optimally during a live echocardiographic examination for detection of the aortic valve. An automated detection system in an echocardiogram will improve the accuracy of medical diagnosis and can provide further medical analysis from the resulting detection. Methods: Two detection architectures, Single Shot Multibox Detector (SSD) and Faster Regional based Convolutional Neural Network (R-CNN) with various feature extractors were trained on echocardiography images from 33 patients. Thereafter, the models were tested on 10 echocardiography videos. Results: Faster R-CNN Inception v2 had shown the highest accuracy (98.6%) followed closely by SSD Mobilenet v2. In terms of speed, SSD Mobilenet v2 resulted in a loss of 46.81% in framesper- second (fps) during real-time detection but managed to perform better than the other neural network models. Additionally, SSD Mobilenet v2 used the least amount of Graphic Processing Unit (GPU) but the Central Processing Unit (CPU) usage was relatively similar throughout all models. Conclusion: Our findings provide a foundation for implementing a convolutional detection system to echocardiography for medical purposes.

scholarly journals An Effective Deep Neural Network Architecture for Cross-Subject Epileptic Seizure Detection in EEG Data

2021 ◽  
Author(s):  
Imene Jemal ◽  
Amar Mitiche ◽  
Lina Abou-Abbas ◽  
Khadidja Henni ◽  
Neila Mezghani
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
False Positive Rate ◽  
Epileptic Seizures ◽  
Seizure Detection ◽  
Model Complexity ◽  
Eeg Data ◽  
Positive Rate ◽  
Active Research ◽  
Public Datasets

For several decades, the detection of epileptic seizures has been an active research topic. The performance of current patient-specific algorithms is satisfactory. However, due to significant variability of EEG data between patients, cross-subject seizure characterization and detection remains a challenging task. The purpose of this study is to propose and investigate a modified convolutional neural network (CNN) architecture based on separable depth-wise convolution for effective automatic cross-subject seizure detection. The architecture is conceived with a reduced number of trainable parameters to reduce the model complexity and storage requirements to easily deploy it in connected devices for real-time seizure detection. The performance of the proposed method is evaluated on two public datasets collected in the Children’s Hospital Boston and the University of Bonn respectively. The method achieves the highest sensitivity-false positive rate/h of 91.93%–0.005, 100%–0.057 for the CHB-MIT and Ubonn datasets respectively.

scholarly journals Artificial Intelligence-Assisted Colonoscopy for Detection of Colon Polyps: a Prospective, Randomized Cohort Study

2020 ◽  
Author(s):  
Yuchen Luo ◽  
Yi Zhang ◽  
Ming Liu ◽  
Yihong Lai ◽  
Panpan Liu ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Real Time ◽  
High Performance ◽  
Detection System ◽  
Random Order ◽  
Colon Polyps ◽  
Clinical Environment ◽  
Polyp Detection ◽  
Link Type ◽  
Polyp Detection Rate

Abstract Background and aims Improving the rate of polyp detection is an important measure to prevent colorectal cancer (CRC). Real-time automatic polyp detection systems, through deep learning methods, can learn and perform specific endoscopic tasks previously performed by endoscopists. The purpose of this study was to explore whether a high-performance, real-time automatic polyp detection system could improve the polyp detection rate (PDR) in the actual clinical environment. Methods The selected patients underwent same-day, back-to-back colonoscopies in a random order, with either traditional colonoscopy or artificial intelligence (AI)-assisted colonoscopy performed first by different experienced endoscopists (> 3000 colonoscopies). The primary outcome was the PDR. It was registered with clinicaltrials.gov. (NCT047126265). Results In this study, we randomized 150 patients. The AI system significantly increased the PDR (34.0% vs 38.7%, p < 0.001). In addition, AI-assisted colonoscopy increased the detection of polyps smaller than 6 mm (69 vs 91, p < 0.001), but no difference was found with regard to larger lesions. Conclusions A real-time automatic polyp detection system can increase the PDR, primarily for diminutive polyps. However, a larger sample size is still needed in the follow-up study to further verify this conclusion. Trial Registration clinicaltrials.gov Identifier: NCT047126265

scholarly journals Robust Approach to Supervised Deep Neural Network Training for Real-Time Object Classification in Cluttered Indoor Environment

2021 ◽  
Vol 11 (15) ◽  
pp. 7148
Author(s):  
Bedada Endale ◽  
Abera Tullu ◽  
Hayoung Shi ◽  
Beom-Soo Kang
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Architecture ◽  
Input Data ◽  
Deep Neural Network ◽  
Data Augmentation ◽  
Object Classification ◽  
Training Data ◽  
Gradient Descent Algorithm

Unmanned aerial vehicles (UAVs) are being widely utilized for various missions: in both civilian and military sectors. Many of these missions demand UAVs to acquire artificial intelligence about the environments they are navigating in. This perception can be realized by training a computing machine to classify objects in the environment. One of the well known machine training approaches is supervised deep learning, which enables a machine to classify objects. However, supervised deep learning comes with huge sacrifice in terms of time and computational resources. Collecting big input data, pre-training processes, such as labeling training data, and the need for a high performance computer for training are some of the challenges that supervised deep learning poses. To address these setbacks, this study proposes mission specific input data augmentation techniques and the design of light-weight deep neural network architecture that is capable of real-time object classification. Semi-direct visual odometry (SVO) data of augmented images are used to train the network for object classification. Ten classes of 10,000 different images in each class were used as input data where 80% were for training the network and the remaining 20% were used for network validation. For the optimization of the designed deep neural network, a sequential gradient descent algorithm was implemented. This algorithm has the advantage of handling redundancy in the data more efficiently than other algorithms.

scholarly journals FetNet: a recurrent convolutional network for occlusion identification in fetoscopic videos

2020 ◽  
Vol 15 (5) ◽  
pp. 791-801 ◽  
Author(s):  
Sophia Bano ◽  
Francisco Vasconcelos ◽  
Emmanuel Vander Poorten ◽  
Tom Vercauteren ◽  
Sebastien Ourselin ◽  
...  
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Laser Photocoagulation ◽  
Short Term Memory ◽  
Total Duration ◽  
Frame Rate ◽  
Superior Performance ◽  
Computer Assisted ◽  
Convolutional Network ◽  
Spatio Temporal

Abstract Purpose Fetoscopic laser photocoagulation is a minimally invasive surgery for the treatment of twin-to-twin transfusion syndrome (TTTS). By using a lens/fibre-optic scope, inserted into the amniotic cavity, the abnormal placental vascular anastomoses are identified and ablated to regulate blood flow to both fetuses. Limited field-of-view, occlusions due to fetus presence and low visibility make it difficult to identify all vascular anastomoses. Automatic computer-assisted techniques may provide better understanding of the anatomical structure during surgery for risk-free laser photocoagulation and may facilitate in improving mosaics from fetoscopic videos. Methods We propose FetNet, a combined convolutional neural network (CNN) and long short-term memory (LSTM) recurrent neural network architecture for the spatio-temporal identification of fetoscopic events. We adapt an existing CNN architecture for spatial feature extraction and integrated it with the LSTM network for end-to-end spatio-temporal inference. We introduce differential learning rates during the model training to effectively utilising the pre-trained CNN weights. This may support computer-assisted interventions (CAI) during fetoscopic laser photocoagulation. Results We perform quantitative evaluation of our method using 7 in vivo fetoscopic videos captured from different human TTTS cases. The total duration of these videos was 5551 s (138,780 frames). To test the robustness of the proposed approach, we perform 7-fold cross-validation where each video is treated as a hold-out or test set and training is performed using the remaining videos. Conclusion FetNet achieved superior performance compared to the existing CNN-based methods and provided improved inference because of the spatio-temporal information modelling. Online testing of FetNet, using a Tesla V100-DGXS-32GB GPU, achieved a frame rate of 114 fps. These results show that our method could potentially provide a real-time solution for CAI and automating occlusion and photocoagulation identification during fetoscopic procedures.

scholarly journals Computer-Aided Colon Polyp Detection on High Resolution Colonoscopy Using Transfer Learning Techniques

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5315
Author(s):  
Chia-Pei Tang ◽  
Kai-Hong Chen ◽  
Tu-Liang Lin
Keyword(s):  
Real Time ◽  
Transfer Learning ◽  
Early Recognition ◽  
Detection System ◽  
Colon Polyp ◽  
Colon Polyps ◽  
Polyp Detection ◽  
Detection Model ◽  
Real Time Detection ◽  
Polyp Type

Colonoscopies reduce the incidence of colorectal cancer through early recognition and resecting of the colon polyps. However, the colon polyp miss detection rate is as high as 26% in conventional colonoscopy. The search for methods to decrease the polyp miss rate is nowadays a paramount task. A number of algorithms or systems have been developed to enhance polyp detection, but few are suitable for real-time detection or classification due to their limited computational ability. Recent studies indicate that the automated colon polyp detection system is developing at an astonishing speed. Real-time detection with classification is still a yet to be explored field. Newer image pattern recognition algorithms with convolutional neuro-network (CNN) transfer learning has shed light on this topic. We proposed a study using real-time colonoscopies with the CNN transfer learning approach. Several multi-class classifiers were trained and mAP ranged from 38% to 49%. Based on an Inception v2 model, a detector adopting a Faster R-CNN was trained. The mAP of the detector was 77%, which was an improvement of 35% compared to the same type of multi-class classifier. Therefore, our results indicated that the polyp detection model could attain a high accuracy, but the polyp type classification still leaves room for improvement.

PAIRWISE CLUSTERS OPTIMIZATION AND CLUSTER MOST SIGNIFICANT FEATURE METHODS FOR ANOMALY-BASED NETWORK INTRUSION DETECTION SYSTEM (POC2MSF)

2018 ◽  
Vol 3 (2) ◽  
pp. 93
Author(s):  
Gervais Hatungimana
Keyword(s):  
Intrusion Detection ◽  
Network Traffic ◽  
False Positive ◽  
Intrusion Detection System ◽  
Detection System ◽  
False Positive Rate ◽  
Significant Feature ◽  
Features Selection ◽  
Number Of Clusters ◽  
Network Intrusion

 Anomaly-based Intrusion Detection System (IDS) uses known baseline to detect patterns which have deviated from normal behavior. If the baseline is faulty, the IDS performance degrades. Most of researches in IDS which use k-centroids-based clustering methods like K-means, K-medoids, Fuzzy, Hierarchical and agglomerative algorithms to baseline network traffic suffer from high false positive rate compared to signature-based IDS, simply because the nature of these algorithms risk to force some network traffic into wrong profiles depending on K number of clusters needed. In this paper we propose alternate method which instead of defining K number of clusters, defines t distance threshold. The unrecognizable IDS; IDS which is neither HIDS nor NIDS is the consequence of using statistical methods for features selection. The speed, memory and accuracy of IDS are affected by inappropriate features reduction method or ignorance of irrelevant features. In this paper we use two-step features selection and Quality Threshold with Optimization methods to design anomaly-based HIDS and NIDS separately. The performance of our system is 0% ,99.9974%, 1,1 false positive rates, accuracy , precision and recall respectively for NIDS and  0%,99.61%, 0.991,0.978 false positive rates, accuracy, precision and recall respectively for HIDS.

scholarly journals Hybrid Internal Anomaly Detection System for IoT: Reactive Nodes with Cross-Layer Operation

2018 ◽  
Vol 2018 ◽  
pp. 1-15 ◽  
Author(s):  
Nanda Kumar Thanigaivelan ◽  
Ethiopia Nigussie ◽  
Seppo Virtanen ◽  
Jouni Isoaho
Keyword(s):  
Anomaly Detection ◽  
False Positive ◽  
Detection System ◽  
False Positive Rate ◽  
Security Analysis ◽  
Abnormal Behavior ◽  
Test Bed ◽  
Positive Rate ◽  
Parent Node ◽  
Anomaly Detection System

We present a hybrid internal anomaly detection system that shares detection tasks between router and nodes. It allows nodes to react instinctively against the anomaly node by enforcing temporary communication ban on it. Each node monitors its own neighbors and if abnormal behavior is detected, the node blocks the packets of the anomaly node at link layer and reports the incident to its parent node. A novel RPL control message, Distress Propagation Object (DPO), is formulated and used for reporting the anomaly and network activities to the parent node and subsequently to the router. The system has configurable profile settings and is able to learn and differentiate between the nodes normal and suspicious activities without a need for prior knowledge. It has different subsystems and operation phases that are distributed in both the nodes and router, which act on data link and network layers. The system uses network fingerprinting to be aware of changes in network topology and approximate threat locations without any assistance from a positioning subsystem. The developed system was evaluated using test-bed consisting of Zolertia nodes and in-house developed PandaBoard based gateway as well as emulation environment of Cooja. The evaluation revealed that the system has low energy consumption overhead and fast response. The system occupies 3.3 KB of ROM and 0.86 KB of RAM for its operations. Security analysis confirms nodes reaction against abnormal nodes and successful detection of packet flooding, selective forwarding, and clone attacks. The system’s false positive rate evaluation demonstrates that the proposed system exhibited 5% to 10% lower false positive rate compared to simple detection system.

Sign in / Sign up

Export Citation Format

Share Document