Design and Implementation of an Intensive Care Unit Command Center for Medical Data Fusion.

The rapid advancements in Artificial Intelligence of Things (AIoT) are pivotal for the healthcare sector, especially as the world approaches an aging society which will be reached by 2050. This paper presents an innovative AIoT-enabled data fusion system implemented at the CMUH Respiratory Intensive Care Unit (RICU) to address the high incidence of medical errors in ICUs, which are among the top three causes of mortality in healthcare facilities. ICU patients are particularly vulnerable to medical errors due to the complexity of their conditions and the critical nature of their care. We introduce a four-layer AIoT architecture designed to manage and deliver both real-time and non-real-time medical data within the CMUH-RICU. Our system demonstrates the capability to handle 22 TB of medical data annually with an average delay of 1.72 ms and a bandwidth of 65.66 Mbps. Additionally, we ensure the uninterrupted operation of the CMUH-RICU with a three-node streaming cluster (called Kafka), provided a failed node is repaired within 9 h, assuming a one-year node lifespan. A case study is presented where the AI application of acute respiratory distress syndrome (ARDS), leveraging our AIoT data fusion approach, significantly improved the medical diagnosis rate from 52.2% to 93.3% and reduced mortality from 56.5% to 39.5%. The results underscore the potential of AIoT in enhancing patient outcomes and operational efficiency in the ICU setting.

An Edge Transfer Learning Approach for Calibrating Soil Electrical Conductivity Sensors.

Smart agriculture utilizes Internet of Things (IoT) technologies to enable low-cost electrical conductivity (EC) sensors to support farming intelligence. Due to aging and changes in weather and soil conditions, EC sensors are prone to long-term drift over years of operation. Therefore, regular recalibration is necessary to ensure data accuracy. In most existing solutions, an EC sensor is calibrated by using the standard sensor to build the calibration table. This paper proposes SensorTalk3, an ensemble approach of machine learning models including XGBOOST and Random Forest, which can be executed at an edge device (e.g., Raspberry Pi) without GPU acceleration. Our study indicates that the soil information (both temperature and moisture sensor data) plays an important role in SensorTalk3, which significantly outperforms the existing calibration approaches. The MAPE of SensorTalk3 can be as low as 1.738%, compared to the 7.792% error of the original sensor. Our study indicates that when the errors of uncalibrated moisture and temperature sensors are not larger than 8.3%, SensorTalk3 can accurately calibrate EC. SensorTalk3 can perform model training during data collection at the edge node. When all training data are collected, AI training is also finished at the edge node. Such an AI training approach has not been found in existing edge AI approaches. We also proposed the dual-sensor detection solution to determine when to conduct recalibration. The overhead of this solution is less than twice the optimal detection scenario (which cannot be achieved practically). If the two non-standard sensors are homogeneous and stable, then the optimal detection scenario can be approached. Conventional methods require training calibration AI models in the cloud. However, SensorTalk3 introduces a significant advancement by enabling on-site transfer learning in the edge node. Given the abundance of farming sensors deployed in the fields, performing local transfer learning using low-cost edge nodes proves to be a more cost-effective solution for farmers.

Enhancing travel time prediction with deep learning on chronological and retrospective time order information of big traffic data.

With growing environmental concerns and the exploitation of ubiquitous big data, smart transportation is transforming logistics business and operations into a more sustainable approach. To answer questions in intelligent transportation planning, such as which data are feasible, which methods are applicable for intelligent prediction of such data, and what are the available operations for prediction, this paper offers a new deep learning approach called bi-directional isometric-gated recurrent unit (BDIGRU). It is merged to the deep learning framework of neural networks for predictive analysis of travel time and business adoption for route planning. The proposed new method directly learns high-level features from big traffic data and reconstructs them by its own attention mechanism drawn by temporal orders to complete the learning process recursively in an end-to-end manner. After deriving the computational algorithm with stochastic gradient descent, we use the proposed method to perform predictive analysis of stochastic travel time under various traffic conditions (especially for congestions) and then determine the optimal vehicle route with the shortest travel time under future uncertainty. Based on empirical results with big traffic data, we show that the proposed BDIGRU method can (1) significantly improve the predictive accuracy of one-step 30 min ahead travel time compared to several conventional (data-driven, model-driven, hybrid, and heuristics) methods measured with several performance criteria, and (2) efficiently determine the optimal vehicle route in relation to the predictive variability under uncertainty.

DeviceTalk: A No-Code Low-Code IoT Device Code Generation.

The deployment of a client-server-based distributed intelligent system involves application development in both the network domain and the device domain. In the network domain, an application server (typically in the cloud) is deployed to execute the network applications. In the device domain, several Internet of Things (IoT) devices may be configured as, for example, wireless sensor networks (WSNs), and interact with each other through the application server. Developing the network and the device applications are tedious tasks that are the major costs for building a distributed intelligent system. To resolve this issue, a low-code or no-code (LCNC) approach has been purposed to automate code generation. As traditional LCNC solutions are highly generic, they tend to generate excess code and instructions, which will lack efficiency in terms of storage and processing. Fortunately, optimization of automated code generation can be achieved for IoT by taking advantage of the IoT characteristics. An IoT-based distributed intelligent system consists of the device domain (IoT devices) and the network domain (IoT server). The software of an IoT device in the device domain consists of the Device Application (DA) and the Sensor Application (SA). Most IoT LCNC approaches provide code generation in the network domain. Very few approaches automatically generate the DA code. To our knowledge, no approach supports the SA code generation. In this paper, we propose DeviceTalk, an LCNC environment for the DA and the SA code development. DeviceTalk automatically generates the code for IoT devices to speed up the software development in the device domain for a distributed intelligent system. We propose the DeviceTalk architecture, design and implementation of the code generation mechanism for the IoT devices. Then, we show how a developer can use the DeviceTalk Graphical User Interface (GUI) to exercise LCNC development of the device software.

Application of deep machine learning for the radiographic diagnosis of periodontitis.

OBJECTIVE: Successful application of deep machine learning could reduce time-consuming and labor-intensive clinical work of calculating the amount of radiographic bone loss (RBL) in diagnosing and treatment planning for periodontitis. This study aimed to test the accuracy of RBL classification by machine learning. MATERIALS AND METHODS: A total of 236 patients with standardized full mouth radiographs were included. Each tooth from the periapical films was evaluated by three calibrated periodontists for categorization of RBL and radiographic defect morphology. Each image was pre-processed and augmented to ensure proper data balancing without data pollution, then a novel multitasking InceptionV3 model was applied. RESULTS: The model demonstrated an average accuracy of 0.87 ± 0.01 in the categorization of mild (< 15%) or severe (≥ 15%) bone loss with fivefold cross-validation. Sensitivity, specificity, positive predictive, and negative predictive values of the model were 0.86 ± 0.03, 0.88 ± 0.03, 0.88 ± 0.03, and 0.86 ± 0.02, respectively. CONCLUSIONS: Application of deep machine learning for the detection of alveolar bone loss yielded promising results in this study. Additional data would be beneficial to enhance model construction and enable better machine learning performance for clinical implementation. CLINICAL RELEVANCE: Higher accuracy of radiographic bone loss classification by machine learning can be achieved with more clinical data and proper model construction for valuable clinical application.

VerificationTalk: A Verification and Security Mechanism for IoT Applications.

An Internet of Things (IoT) application typically involves implementations in both the device domain and the network domain. In this two-domain environment, it is possible that application developers implement the wrong network functions and/or connect some IoT devices that should never be linked, which result in the execution of wrong operations on network functions. To resolve these issues, we propose the VerificationTalk mechanism to prevent inappropriate IoT application deployment. VerificationTalk consists of two subsystems: the BigraphTalk subsystem which verifies IoT device configuration; and AFLtalk which validates the network functions. VerificationTalk provides mechanisms to conduct online anomaly detection by using a runtime monitor and offline by using American Fuzzy Lop (AFL). The runtime monitor is capable of intercepting potentially harmful data targeting IoT devices. When VerificationTalk detects errors, it provides feedback for debugging. VerificationTalk also assists in building secure IoT applications by identifying security loopholes in network applications. By the appropriate design of the IoTtalk execution engine, the testing capacity of AFLtalk is three times that of traditional AFL approaches.

SpecTalk: Conforming IoT Implementations to Sensor Specifications.

Due to the fast evolution of Sensor and Internet of Things (IoT) technologies, several large-scale smart city applications have been commercially developed in recent years. In these developments, the contracts are often disputed in the acceptance due to the fact that the contract specification is not clear, resulting in a great deal of discussion of the gray area. Such disputes often occur in the acceptance processes of smart buildings, mainly because most intelligent building systems are expensive and the operations of the sub-systems are very complex. This paper proposes SpecTalk, a platform that automatically generates the code to conform IoT applications to the Taiwan Association of Information and Communication Standards (TAICS) specifications. SpecTalk generates a program to accommodate the application programming interface of the IoT devices under test (DUTs). Then, the devices can be tested by SpecTalk following the TAICS data formats. We describe three types of tests: self-test, mutual-test, and visual test. A self-test involves the sensors and the actuators of the same DUT. A mutual-test involves the sensors and the actuators of different DUTs. A visual-test uses a monitoring camera to investigate the actuators of multiple DUTs. We conducted these types of tests in commercially deployed applications of smart campus constructions. Our experiments in the tests proved that SpecTalk is feasible and can effectively conform IoT implementations to TACIS specifications. We also propose a simple analytic model to select the frequency of the control signals for the input patterns in a SpecTalk test. Our study indicates that it is appropriate to select the control signal frequency, such that the inter-arrival time between two control signals is larger than 10 times the activation delay of the DUT.

AMBtalk: A Cardiovascular IoT Device for Ambulance Applications.

Acute Coronary Syndrome (ACS) and other heart emergency events require immediate chest pain identification in the ambulance. Specifically, early identification and triage is required so that patients with chest pain can be quickly sent to a hospital with appropriate care facilities for treatment. In the traditional approach, ambulance personnel often use symptom checklists to examine the patient and make a quick decision for the target hospital. However, not every hospital has specialist facilities to handle such emergency cases. If the result of the subsequent cardiac enzyme test performed at the target hospital strongly suggests the occurrence of myocardial infarction, the patient may need to be sent to another hospital with specialist facilities, such as Percutaneous Coronary Intervention. The standard procedure is time consuming, which may result in delayed treatment and reduce patent survival rate. To resolve this issue, we propose AMBtalk (Ambulance Talk) for accurate, early ACS identification in an ambulance. AMBtalk provides real-time connection to hospital resources, which reduces the elapsed time for treatment, and therefore, improves the patient survival rate. The key to success for AMBtalk is the development of the AllCheck® Internet of Things (IoT) device, which can accurately and quickly provide cardiovascular parameter values for early ACS identification. The interactions between the AllCheck® IoT device, the emergency medical service center, the ambulance personnel and the hospital are achieved through the AMBtalk IoT server in the cloud network. AllCheck® outperforms the existing cardiovascular IoT device solutions for ambulance applications. The testing results of the AllCheck® device show 99% correlation with the results of the hospital reports. Due to its excellent performance in quick ACS identification, the AllCheck® device was awarded the 17th Taiwan Innovators Award in 2020.

SimTalk: Simulation of IoT Applications.

The correct implementation and behavior of Internet of Things (IoT) applications are seldom investigated in the literature. This paper shows how the simulation mechanism can be integrated well into an IoT application development platform for correct implementation and behavior investigation. We use an IoT application development platform called IoTtalk as an example to describe how the simulation mechanism called SimTalk can be built into this IoT platform. We first elaborate on how to implement the simulator for an input IoT device (a sensor). Then we describe how an output IoT device (an actuator) can be simulated by an animated simulator. We use a smart farm application to show how the simulated sensors are used for correct implementation. We use applications including interactive art (skeleton art and water dance) and the pendulum physics experiment as examples to illustrate how IoT application behavior investigation can be achieved in SimTalk. As the main outcome of this paper, the SimTalk simulation codes can be directly reused for real IoT applications. Furthermore, SimTalk is integrated well with an IoT application verification tool in order to formally verify the IoT application configuration. Such features have not been found in any IoT simulators in the world.

SensorTalk: An IoT Device Failure Detection and Calibration Mechanism for Smart Farming.

In an Internet of Things (IoT) system, it is essential that the data measured from the sensors are accurate so that the produced results are meaningful. For example, in AgriTalk, a smart farm platform for soil cultivation with a large number of sensors, the produced sensor data are used in several Artificial Intelligence (AI) models to provide precise farming for soil microbiome and fertility, disease regulation, irrigation regulation, and pest regulation. It is important that the sensor data are correctly used in AI modeling. Unfortunately, no sensor is perfect. Even for the sensors manufactured from the same factory, they may yield different readings. This paper proposes a solution called SensorTalk to automatically detect potential sensor failures and calibrate the aging sensors semi-automatically. Numerical examples are given to show the calibration tables for temperature and humidity sensors. When the sensors control the actuators, the SensorTalk solution can also detect whether a failure occurs within a detection delay. Both analytic and simulation models are proposed to appropriately select the detection delay so that, when a potential failure occurs, it is detected reasonably early without incurring too many false alarms. Specifically, our selection can limit the false detection probability to be less than 0.7%.

PlantTalk: A Smartphone-Based Intelligent Hydroponic Plant Box.

This paper proposes an IoT-based intelligent hydroponic plant factory solution called PlantTalk. The novelty of our approach is that the PlantTalk intelligence can be built through an arbitrary smartphone. We show that PlantTalk can flexibly configure the connections of various plant sensors and actuators through a smartphone. One can also conveniently write Python programs for plant-care intelligence through the smart phone. The developed plant-care intelligence includes automatic LED lighting, water spray, water pump and so on. As an example, we show that the PlantTalk intelligence effectively lowers the CO2 concentration, and the reduction speed is 53% faster than a traditional plant system. PlantTalk has been extended for a plant factory called AgriTalk.

The SDN Approach for the Aggregation/Disaggregation of Sensor Data.

In many Internet of Things (IoT) applications, large numbers of small sensor data are delivered in the network, which may cause heavy traffics. To reduce the number of messages delivered from the sensor devices to the IoT server, a promising approach is to aggregate several small IoT messages into a large packet before they are delivered through the network. When the packets arrive at the destination, they are disaggregated into the original IoT messages. In the existing solutions, packet aggregation/disaggregation is performed by software at the server, which results in long delays and low throughputs. To resolve the above issue, this paper utilizes the programmable Software Defined Networking (SDN) switch to program quick packet aggregation and disaggregation. Specifically, we consider the Programming Protocol-Independent Packet Processor (P4) technology. We design and develop novel P4 programs for aggregation and disaggregation in commercial P4 switches. Our study indicates that packet aggregation can be achieved in a P4 switch with its line rate (without extra packet processing cost). On the other hand, to disaggregate a packet that combines N IoT messages, the processing time is about the same as processing N individual IoT messages. Our implementation conducts IoT message aggregation at the highest bit rate (100 Gbps) that has not been found in the literature. We further propose to provide a small buffer in the P4 switch to significantly reduce the processing power for disaggregating a packet.

A novel fluorescent biosensor for detection of target DNA fragment from the transgene cauliflower mosaic virus 35S promoter.

In this paper, we reported a convenient fluorescence method for the detection of genetically modified organisms (GMOs). As it is known that the cauliflower mosaic virus (CaMV) 35S promoter is widely used in most transgenic plants (Schnurr and Guerra, 2000), we thus design a simple method based on the detection of a section target DNA (DNA-T) from the transgene CaMV 35S promoter. In this method, the full-length guanine-rich single-strand sequences were split into fragments (Probe 1 and 2) and each part of the fragment possesses two GGG repeats. In the presence of K(+) ion and berberine, if a complementary target DNA of the CaMV 35S promoter was introduced to hybridize with Probe 1 and 2, a G-quadruplex-berberine complex was thus formed and generated a strong fluorescence signal. The generation of fluorescence signal indicates the presence of CaMV 35S promoter. This method is able to identify and quantify Genetically Modified Organisms (GMOs), and it shows wide linear ranges from 5.0×10(-9) to 9.0×10(-7) mol/L with a detection limit of 2.0×10(-9) mol/L.