The network technology plays an important role in the industrial control systems (ICS)then has become the target of cyber-attackers. For industrial internet-of-things (IIoT) applications with limited computing resources, designing an effective NIDS is challenging. A compacteffective NIDS for IIoT is proposedvalidated by using the more recent UNSW-NB 15 dataset to improve the detection capability against new types of attacks in the real world. Experimental results show that the proposed method achieves better performance than previous methods.

By collecting the changes of kinds of vital signs(including heart beat / blood pressure / steps), we combined the results with the testers’ fatigue scores which belonged to the matched fatigue typestheir basic profiles. We extracted 780 types of structural features sets by cleaning raw data, then separating into training set, validation settesting set. With machine learning algorithm to auto-optimize regression analysis, finally we come out the correlative features between vital signsfatigue ,the overwork model as our deliverables.

As we continue to face the rapid increase in confirmed Coronavirus cases around the world, we created an AI-based pneumonia detection platform for COVID-19. The system is able to automatically detect high-risk patients with pneumonia that will then send information to doctors. With that information, the doctors are then able to make follow-up decisions and provide a treatment plan after the diagnosis. In specific, doctors from the Department of Medical Imaging provided us thousands of positive and negative chest x-rays for pneumonia as a training set. Our system has already been tested with and adopted by doctors at the NCKU Hospital. The system achieved 95% accuracy to detect the pneumonia symptom, based on 1400 test images.

Use NLP techniques to realize the automatic coding of ICD10. According to the input of the patient’s age, gender, medical order, admissions, progress note, surgical records, discharge, ICD-10 diagnostic codeICD-10 disposal code, perform machine learning model trainingcode prediction. In addition, the combination codemedical order-related coding rules in practice are used to establish corresponding rules to optimize the accuracy of AI prediction.

With the rise in computing power, deep-learning based computer-aided diagnosis systems have gained interest in the research community. Our system process the images to assist doctors to determine whether the patients have nodules in lungs. Meanwhile, we utilized the Feature Pyramid Network to extend the receptive field on the convolutional kernel, which improved the performance on the nodule detection with various locations in CXR. The semi-supervised learning mechanism also achieves the way of soft-annotation to reduce human effort in medical image annotation.

Based on our long-term accumulated healthy brain image database, we used meta-analysismultivariate analytical algorithms to extract the cognitive biological characteristics of individual subjects. By using machine learning / artificial intelligence algorithms, these features were further calculated for capturing individual cognition according to related brain networks. This biological index could be a potential marker for evaluating individual brain health, cognitive functions,risk for neurodegenerative diseases.

"The primary goal of this project is to establish a complete hospital-based liver cancer database, profiles for data feature extraction,develop different cancer, prediction models. A Medical AI program to predict the poster treatment (including operationradiofrequency ablation) recurrence of liver cancer will be established. The program system will assist doctors in the medical decision, identify high-risk patients,adjust clinical follow-up programs."

The interpretation of next generation sequencing is a big challenge. In order to improve the molecular diagnosis of the patient, we developed the AI_Variant prioritizer, a machine learning based variant prioterization system that can help to prioritize candidate variants for human disease. This module highlighted the possible disease causing variants to help clinician to increase the disease diagnostic yield and decrease the load of manpower.

Ted-ICU AI Platform: (1)Provide a single view of patients＇ EMRsvital signs (2)Support remote ICU monitoring (3)Online labeling tool & Expandable AI algorithms repository (4)Disease-specific prediction models (5)Standardized EMR templates

This technology integrates 1000+ times of UAV imaging experiences with labeled rice lodging images for training. A rice lodging recognition model using deep learning reaches 90 accuracy. The recognition model can be deployed in a microcomputer mounted on UAVs to implement edge computing. While taking aerial images, the inference can be completedreveal lodging areadamage level in-time.

This technique uses Tainan 11 (TN11), which occupies about 70 of the rice planting area in Taiwan. Building the early watering warning system of crops by combining the physiological indicators of rice with environmental dataartificial intelligence technology. The system can analyzepredict the impending water shortage for rice in advanceissue drought warning immediately. Therefore, decision makers can give the timelyappropriate amount for rice irrigationimprove the efficiency of using waterreducing the cost of farmersfarming losses.

In the context of FinTech, we present Financial Graph Attention Networks (FinGAT) to recommend high-profitable stocks in terms of return ratio using time series of stock pricessector info. Our FinGAT can learn the long-short-term price tendency,model the latent interactionsinfluence between stockssectors without any hand-crafted effort. Experiments conducted on Taiwan Stock, S&P 500,NASDAQ stock markets exhibit remarkable accuracy of FinGAT, comparing to state-of-the-arts (by 11, 14,12 performance improvement). The tech is published in IEEE TKDE 2021.

This system uses the innovative AIoT application to target annoying scalp maintenance problemsdevelops an intelligent scalp detectionmanagement system. The core of the technology is to use deep learning-based object detection models. Tens of thousands of microscopic image data sets that are labeledtrained for scalp symptoms. As a result, the scalp image recognition module is develop, such as dandruff, hair loss, oil, and inflammation, etc. Hence this system can provide scalp detection, and maintenance effectiveness tracking functions lead scalp maintenance services to a new level of intelligent management.

We apply diffuse reflectance spectroscopy method to accurately determine the skin properties of subjects. Our device has a very simple configurationthe detected reflectance is processed with an advanced, efficient machine learning algorithm to quantify the skin composition of subjects.

With the astronomically growing trade flows, customs administrations need effective and explainable methods to detect suspicious transactions. This project presents a novel artificial intelligence-based model named DATE that ranks trade flows in the order of fraud risk and to maximize customs revenue. We confirm the superiority of DATE over state-of-the-art AI models, with a remarkable precision of 92.7% on illegal cases and a recall of 49.3% on revenue after inspecting only 1% of all trade flows. Predictions of DATE are also interpretable from the attention mechanism. We are deploying DATE in Nigeria and Malawi Customs Services, in collaboration with the World Customs Organization (WCO). DATE has been published in ACM KDD 2020, which is an AI top conference.

The Al-based web diagnostic system provides an online assessment tool for diagnosing schizophrenia. The Explainable Deep Neural Network classifier is deployed to analyze gray matterwhite matter to derive diagnostic classification of schizophrenia. The structural brain abnormalities associated with schizophrenia is visualized on the AI-based web diagnostic system at individual level.

We propose a spatial-temporal inference model, which uses a large amount of spatialtemporal data in the city to help governmentsenterprises predict future long-short-term important urban indicator values, such as traffic flow, human mobility, pollution level, number of criminal caseseven commercial profitability. The SIM model exploits IoT to integrate multiple real-time geospatial big data, including population, the flow of people, geographical data, Traffic,real-time sensor values. SIM can make effective predictionsprovide explainability for making decisions.

The core concept of the decision support system is to establish machine learning-based predictive modelsenable their interpretability to support physicians making clinical decisions in practice. The developed technique is part of the capstone project named "Smart Emergency Department," sponsored by the Ministry of ScienceTechnology. The system comprises NTUH (National Taiwan University Hospital) EMR (Electronic Medical Record) importance analysis, accurate clinical quality indicator prediction,similar medical record retrieval. It is expected to improve the patients＇ flowalleviate emergency department crowding. It has been verified in retrospective studieswill officially enter the clinical trial phase this year.

Integrating deep learning, 3D information analysis, hyperspectral analysis, computer vision analysiscombined the multi-source images to develop fruit quality monitoring techniques, including: planting area monitoring, position monitoring, yield monitoring, harvesting time prediction, fruit maturity measurement, fruit quality testing, to achieve the goal of enhancing industrial value.

This technique employs AIbig data analytics to precisely forecast cooling load demandestimate efficiency of different chiller combinations. Time-of-Use Pricingoptimal chiller load interval are also considered for practical needs. Decision supports of optimal chiller configuration are provided to enhance energy conservation.

This technique combines more than one AI models to precisely predict current under noisy data from current sensoroptimizes speed rate based on this AI forecasting model. At the same time, it considers the practical requirementslimitations of workpiece surface roughness, tool machine current load, etc. This technique provides decision supports for optimizing parameters of CNC milling machine to keep high efficiencyenhance energy conservation.

Through the study of structural biological materialthe multiscale analysissimulation, we construct an initiative integrated platform, which provides user-friendly experience, automatically generates designs of lightweight structure for the demandsoffers various choices of materialssuggestions. This design platform contains great potential for engineering applications.

Constructing a functional connectome and its computational model is a crucial step toward understanding the mechanisms of brain functions. To achieve this goal, we developed two correlated technologies: (1) An all-optical physiology (AOP) that is capable of millisecond volumetric imaging and accurate stimulation in living animal brains. This system allows us to establish functional connectome and neural coding with a single-cell resolution. (2) A cellular-level spiking neural circuit simulation system that is capable of tuning itself based on the input data from the AOP system. We have demonstrated our technologies in the Drosophila late visual system and will apply them in the brains of larger species such as mice. Our technologies will greatly enhance knowledge of brain operation.

With the rapid growth of the aging population, the safety of the hospitalized elderly has become an important issue. Therefore, our team propose an innovative bed-exit alarm system using multiple sensors for bed-exit detection. With data fusion, IoTAI technologies, the system can achieve early detectionsignificantly reduce false alarms. In addition, it can realize real-time video streami

We use features such as asymmetric expression and crooked eyes to assess the risk of facial stroke. Observing the micro vibration of the head caused by the contraction of the heart, and develop a zero-contact facial heart rate and respiration rate detection technology in conjunction with the camera. The technology can accurately measure heart rate and respiration rate in real time, thereby reducing the risk of infection. This technology has obtained two ROC patents (M590433, I689285), two US patents (HEART RATE DETECTION METHOD AND DEVICE THEREOF,MOUTH AND NOSE OCCLUDED DETECTING METHOD AND SYSTEM THEREOF). The possibility of detecting strokes through AI machine learning methods is not only accurate, but also find out signs of stroke early to grasp the best time to seek medical treatment.

We develop a vulnerability detection module to identify three important security vulnerabilities in the IMS system that supports 4G/5G multimedia services. We have validated them in 4 carriers from TaiwanU.S. using smartphones from 7 brands,shown that they can be exploited to launch a stealthy call DoS attack, a social engineering attack with ghost calls,a call ID spoofing attack. Moreover, attackers can detect attackable users using an AI technique with only the users＇ phone numbers. We finally propose solutionsdevelop an AI-based technique of call ID spoofing detection.

We develop an asynchronous framework across userkernel spaces for deep learning applications on the improvement of Wi-Fi performanceimplement it in the driver of commodity Intel Wi-Fi cards. Under this framework, we apply deep reinforcement learning to developing an intelligent rate adaptation (RA) algorithm (DRL-RA), which can achieve the highest throughput in varying channel conditions given many rate options of current Wi-Fi technologies. Its on-learning capability can learn how to efficiently approach the best rate from the experiences of its common usage patternenvironment.

We constructed a model to differentiate different neurodegenerative diseases by using blood-based biomarkers. Linear Discriminant Analysis (LDA) was used to reduce dimensionsMICE (Multivariate Imputation by Chained Equations) was used for missing value treatmentchoose CART model to predict missing value imputation. This machine learning model would be used for early dectection of neurodegenerative diseases.

Due to the fast development of 5G networks, it is critical to identify users service types to allocate resources intelligently. Our technology focuses on users mobility type identification by extracting practical features from users cellular information. We proposed a system architecturehave collected 700-hour data with 150 GB. By using our dataother datasets in the world, we show that our technology can achieve 95 accuracy,reduce 16 energy consumption compared to traditional methods.

As for the proposed technology by combining the artificial bee colony algorithmthe semi-supervised extreme learning machine (ABC-SSELM), characteristic parameter normalization equations of I-V curves are tuned via low-cost data under normal operation of PV strings. The proposed ABC-SSELM method only needs 1-3 labeled data of the total dataset to save humantime cost. The accuracy of diagnosing various mixed faults can reach more than 99.84,the monitoring of dust accumulation can provide effective cleaning to increase the revenue of the solar PV power generation system.