It's common for hypertension patients, especially those over the age of 60, to have the habit of measuring their blood pressure However, using inflatable cuffs can be quite inconvenient From now on, there is a more convenient choice for hypertension patients Maisense's smart, sleeveless blood pressure monitor, making measuring blood pressure no longer a cumbersome task 'Pulse' is an important physiological signal Established in June 2012, Maisense Inc specializes in designing 'mobile medical electronic devicesservices' Its first product, the 'Freescan smart sleeveless blood pressure monitor', is the first in the US to measure blood pressure directly from the wrist pulse without the need for an inflatable cuff It is a medical-grade product and has obtained the EU CE certification for medical materials Blood pressure measured directly from the wrist pulse 'Mai' in Maisense is derived from the Chinese word for 'pulse', which is an important physiological signal in both Western and Chinese medicine Amidst an aging population trend, the medical industry is viewed as very promising for the future, however, it remains one of the few basic industries that has not been fully modernized through internet technology Compared to many industries that strive to please consumers such as retail and banking, the traditional hospital-centered medical structure provides little choice for patients Now, with the trend centered around hospitals changing, the medical industry is facing a new paradigm shift The widespread use of smartphones makes personalized, patient-centric medical care practically feasible, and this type of care is the most effective way to address lifestyle diseases The world's first cuffless medical device - measures blood pressure in just ten seconds As societies age, cardiovascular diseases are not only the leading cause of death worldwide but have also continuously ranked among the top two or three causes of death locally for several years According to statistics from the National Health Department, nearly ten thousand people suffer from cardiovascular-related diseases each year, posing a serious threat to public health and a disease not to be taken lightly in modern society However, the golden rescue time for cardiovascular diseases is very short, needing accurate diagnosis and the right measures to reduce the mortality rates 10-second blood pressure measurement diagram Maisense's Freescan is the world's first cuffless medical device capable of measuring blood pressure in just ten seconds Paired with myFreescan Apps, it quickly measures cardiovascular-related stroke factors, with atrial fibrillation AFib sensitivity reaching up to 967 The device integrates multiple cardiovascular monitoring functions such as blood pressure and electrocardiograms myFreescan health management page Each measurement result can be synchronized with mobile devices, instantly revealing more cardiovascular parameters like atrial fibrillation, with the help of AI technology which assists in analyzing personal data to keep individuals informed about their health status and progress Subsequently, doctors can provide recommendations based on AI-analyzed results when necessary, ensuring an added layer of health protection Data synchronization with mobile device illustration The Freescan product has evolved from a sleeveless blood pressure monitor to a mobile cardiovascular monitoring instrument It can measure key predictive factors for stroke such as blood pressure, atrial fibrillation, arrhythmia, and arterial elasticity without the need for an inflatable cuff This effectively helps to prevent the incidence of strokes and provides the elderly with a more convenient and comfortable option「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

On January 23, 2020, Wuhan, the birthplace of the new coronavirus commonly known as Wuhan pneumonia, was closed On January 27, Taiwan’s first new coronavirus epidemic chatbot platform was established, providing immediate information to the government, NGOs and small and medium-sized enterprises with less than 30 people Free to use to stay informed Through the combination of APIs, information can be automatically updated to ensure that when users want to know information about the epidemic, they can obtain the latest and correct epidemic information, respond to the development of the epidemic early, and effectively reduce the spread of infection In order to combat the new coronavirus epidemic, software giant Microsoft Microsoft is currently working with the Centers for Disease Control and Prevention CDC and some large healthcare providers to deploy a chatbot to help In response to the COVID-19 virus, GoSky has also joined Facebook's global government cooperation program to provide chatbot construction services to assist government public health departments and other units in fighting the epidemic Global chatbot output value will reach US125 billion in 2025 GoSky has observed that many companies have also been greatly affected by the epidemic In order to prevent the epidemic, most people and companies have implemented home-based epidemic prevention and remote working policies Now that offline sales have dropped significantly, GoSky has independently expanded the epidemic prevention project to small and medium-sized enterprises with less than 30 employees According to a Grand View Research report, the global chatbot market will grow at a compound growth rate of 243 from 2017 to 2025, reaching an output value of US125 billion in 2025 In addition, according to estimates by Gartner, an international research consultancy, by 2021, more than 50 of companies will invest more in chatbots Chatbots than traditional apps each year Chatbot technology is becoming more and more mature, and GoSky believes that Chatbot is the next app Chatbot will be the next app Chat robot diagram GoSky, a technology company founded in March 2018, is the world's largest chat robot platform More than 46 of the world's chat robots are built on its platform, providing one-stop chat robots and digital marketing Weiquanlong, Yanik, Born Night Show, Uber Eats, etc are all its customers, and the service industry covers the news media, department store industry, express delivery industry, etc At the end of 2019, Wuhan pneumonia new coronavirus broke out in China The Taiwan government announced the establishment of a central epidemic command center at the end of December 2020 and screened overseas passengers GoSky realized that "the spread of epidemic information is faster than the spread of the epidemic, and there is an opportunity to significantly reduce the impact of the epidemic" After a quick discussion with the team, GoSky decided to use the technology and experience of making chatbots in the past to integrate Johns Hopkins University, USA The CSSE Center for Systems Science and Engineering at Hopkins University opened the source code of the Global Epidemic Information System, WHO, the US CDC, the Taiwan Disease Control and Prevention Agency and other agencies released epidemic data On January 27, 2020, the third day of the Lunar New Year, it took a day In the evening, we deployed ahead of schedule to complete the production of the Wuhan Pneumonia Chatbot Chatbot helps control the global epidemic GoSky Marketing Manager Jiang Yanzhou said that the Wuhan Pneumonia Chatbot initially has several functions 1 Real-time data on the global epidemic, including the number of confirmed, recovered, and dead people around the world statistics on the number of confirmed, recovered, and dead people in Taiwan 2 Combined with Google search engine , query real-time news related to Wuhan pneumonia, integrating the latest news from Wikipedia, BBC News Network, China Dragon TV and major domestic news channels 3 One-click dial 1922 epidemic prevention hotline, providing a more direct inquiry channel and providing users with faster Contact the Centers for Disease Control 4 Add it to the desktop with one click, allowing users to follow the development of the epidemic at any time without downloading or occupying mobile phone capacity In the past few weeks, if you want to track the current situation of the epidemic, you have to search through Google, Information can only be obtained by focusing on comprehensive comparisons of major news However, through this chat machine, users can grasp the latest situation anytime, anywhere and in real time, which has a positive impact on the public's understanding of the epidemic New Coronavirus Information Page In the second phase in early February, we joined in the promotion of public health knowledge and cooperated with ICU doctor Dr Chen Zhijin to make the correct knowledge about the use of masks into a lazy bag and added it to the chat robot to help the rapid dissemination of correct public health knowledge At the same time, a mask map was established After the government announced the real-name system for purchasing masks, the open database of the National Health Insurance Bureau was quickly integrated, and information from local health insurance cooperative pharmacies was made into a mask map function Users can click to check the county, city and region to see the real-time mask inventory status in the area Click the button below to open Google Map for users to check the location of the pharmacy for convenient purchase Jiang Yanzhou pointed out that the cloud update feature of the chatbot makes information collection and transmission faster Through the combination of APIs, information can be automatically updated to ensure that when users want to know information related to the epidemic, they can get it The latest and correct epidemic information can respond to the development of the epidemic early and effectively reduce the spread of infection Compared with previous apps, chatbots are built on social communication software There is no need to download additional mobile apps, which greatly reduces the entry barrier for users You can obtain chatbot services by clicking on the link or scanning the QR Code Correct information can also be integrated into the chatbot, allowing users to search at any time At a time when the epidemic information is abundant and complex, it can effectively reduce the impact of false news and ensure the effective delivery of correct information Solution Understand the global COVID-19 epidemic in one second, chatbots become a weapon in the fight against the epidemic New Coronavirus Global Epidemic Information Page NLP technology provides a better interactive experience GoSky is committed to NLP natural semantic analysis technology and builds a database to record user preferences Through machine learning, it understands each user's interaction mode, provides a better interactive process experience, and can also provide customized services for consumers personal service The basic functions of the chatbot are customer service and automatic replies It can share the workload of brand fan page managers editors and help consumers obtain relevant business information more quickly As the new coronavirus continues to spread, according to statistics released by Facebook, the number of messages in areas with severe epidemics has increased by 50 compared with the same period, and users' usage time has doubled This also means that offline During the extraordinary period of declining sales, consumers' demand for online information has increased Through chatbots, customer lists can be collected, allowing companies to better understand consumer preferences and related information, and helping companies quickly resume operations after the epidemic In the future, GoSky will take advantage of chat robots, integrate software and hardware equipment, and develop in the direction of AIoT It is estimated that AIoT products will be launched in mid-2020, so that chat robots will no longer be limited to mobile phone applications Chatbots built on Messenger, which do not occupy mobile phone capacity, are updated in the cloud, and have high opening rates, will be the focus of enterprise development GoSky will focus on the all-round application of chat robots and strive to create a mobile life application ecosystem GoSky team builds a chatbot for the new coronavirus epidemic 「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

As the novel coronavirus epidemic continues to spread, the number of quarantined and isolated individuals steadily increases, becoming an 'unbearable weight' for managers SITU Technology, along with Myte Electronics, Ai Micro Science, and String Cloud Technology, have collectively developed a 'Home Isolation Personal Location Tracking System' for COVID-19, addressing the issues of isolated individuals wandering and holes in epidemic tracking On the afternoon of February 12, 2020, as the novel coronavirus epidemic continued unabated, home-isolated individuals were seen wandering around, causing widespread alarm, with anxious faces hidden beneath white masks A 'Public Cloud Location and Monitoring Seminar' held in Taipei unexpectedly helped forge a significant epidemic prevention coalition Formed by a coalition of established and startup companies including SITU Technology, Myte Electronics, Ai Micro Science, and String Cloud Technology, they introduced the COVID-19 'Home Isolation Personal Location Tracking System' In the abodes of those quarantined, whether in centralized isolation spots or home setups with temperature monitoring platforms, an automated reporting process alerts immediately if abnormalities are detected, significantly reducing the burden on epidemic staff and preventing quarantined individuals from easily leaving their premises, thus plugging the gaps in epidemic tracking Home Isolation Temperature Care System Architecture Diagram Home Isolation Wandering, Epidemic Phones Not Enough As the number of confirmed cases of the coronavirus rises, in order to effectively prevent community transmission, the Central Epidemic Command Center has specified that all entrants must undergo individual home isolation, home quarantine, and self-health management those in home isolation and quarantine must rest at home for 14 days and must not go out, during which community leaders and police officers should ensure at least two temperature checks daily However, as the epidemic intensifies and the number of people requiring a 14-day home quarantine skyrockets, there is a dire shortage of equipment and personnel This includes 2,000 location monitoring phones provided by the government being insufficient additionally, the limited number of epidemic personnel face increasing work pressure, urgently necessitating the use of technological tools to aid the situation Shih-Juan Lin, founder and CEO of SITU Technology, explains that the COVID-19 'Home Isolation Personal Location Tracking System' is divided into hardware and software components The hardware includes a temperature patch from Ai Micro Science and an unlimited location tracking device USB receiver provided by String Cloud Technology The patch, which must be placed in the armpit of the quarantined individual, is capable of measuring temperature 24 hours a day the receiver uploads temperature and location data to the SITU Technology's cloud positioning engine This data is analyzed to track temperature trends and accurately predicts the likelihood of diagnosis Home Isolation Temperature Care System Comparison Chart Temperature Measurement Indoor Positioning System Reduces the Likelihood of Community Infections SITU Technology was officially established in 2015, with co-founders from various fields including indoor positioning, robotic platforms, imaging graphics, machine learning, and cloud computing Shih-Juan Lin states that the unified epidemic phone GPS systems, whether used for centralized quarantine, centralized isolation, home quarantine, or home isolation, have two drawbacks the resolution is too high and they cannot track vertical movements, meaning that they cannot detect movements from one floor to another, or from one room to another, which greatly increases the potential for community transmission Compared to traditional methods, SITU Technology's positioning sensing system uses various devices, including optical lenses, LIDAR, and ultrasound, to sense wireless signals at each point like an indoor Google car, accumulating extensive and comprehensive data to build an accurate floor plan Simply said, by crossing-comparing wireless signals received by multiple indoor receivers and using AI deep learning, the positioning can break through spatial restrictions and achieve greater accuracy In addition to precise positioning, by analyzing the mobility tracks of isolated individuals and groups and tracking their contact history, it is further possible to grasp the virus transmission routes, offer early warnings, and avoid community infections On the management side, computers and large electronic displays can observe the temperature, battery usage, and signal strength of the group patches isolated individuals Once it is detected that an isolator has left their designated location, an automatic alarm signal is immediately triggered, allowing management personnel to take immediate action On the other hand, isolators can also record their own temperature changes on their phones, and in the event of an abnormal situation, they can immediately notify the quarantine unit for further checks Epidemic Platform Diagram This system was conceived and formed on February 12, and within just over a month, has been empirically tested in centralized isolation facilities in Taipei City and New Taipei City of around 50-bed scale, with effective implementation In the future, there is the potential to expand to factories, offices, and other larger scale, high-risk areas for temperature management After the epidemic, this system can also be expanded for use in emergency rooms, maternity wards, postpartum centers, and long-term care institutions Isolator's Temperature Management Interface Amid the ongoing global epidemic, the COVID-19 'Home Isolation Personal Location Tracking System' may have the opportunity to be exported overseas, becoming a global market aide for home isolation and medical quarantine SITU Technology Team「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

Combatting the epidemic is like fighting a fire With the increasing influx of returning residents, the pressure of Wuhan virus quarantine is mounting, consuming more time Shortening quarantine times will positively benefit disease prevention efforts At a hospital in Southern Taiwan, the 'Smart Medical Clinical Decision Support System' has helped reduce the time taken for high-risk patients from entry at the quarantine station to clinical decisions by doctors It originally took about 2 and a half hours, but now it's down to less than 30 minutes, increasing the quarantine efficiency by 5 to 6 times and significantly reducing the risk of cross-infection between medical personnel and patients, as well as the manpower needed for quarantine As waves of overseas students densely return to Taiwan, not only the Central Epidemic Command Center, but also various medical institutions are tightening up, closely monitoring every quarantined individual There are also concerns about the potential infection risks to colleagues, which is exhausting At this point, employing AI technology to enhance quarantine efficiency is indeed a great blessing for the medical units and the health of the nation AI-Assisted Medicine Multiplies, Becoming a Hero in Pandemic Control To combat the severe pandemic of novel coronavirus, the hospital has integrated various smart medical techs and developed the 'Smart Medical Clinical Decision Support System', raising quarantine efficiency by 5 to 6 times It has shortened the time taken for high-risk patients from entering the quarantine station to doctors making clinical decisions from 2 and a half hours to less than 30 minutes, effectively reducing the risk of cross-infections The 'Smart Medical Clinical Decision Support System' implemented by the hospital includes three components front-end automation of medical records, AI-assisted interpretation of chest X-rays for diagnosing pneumonia, and continuous updates of clinical decisions based on the latest epidemic data provided by the health department This significantly enhances the hospital's response and decision-making ability in quarantine and epidemic prevention, and greatly benefits Taiwan's anti-epidemic efforts through the multiplicative effect of AI-assisted medicine National Cheng Kung University collaborates with the hospital, using smart medical technology to enhance quarantine efficiency Photo source Official website In the aspect of medical record automation, existing medical institutions often use traditional paper or verbal reporting, which potentially increases the risk of contact infections among medical staff and patients The automated medical records system in this hospital allows patients to fill out their own medical history, including travel, occupation, contacts, and clustering, using tablet computers These records are uploaded to the electronic medical record system, enabling immediate access by medical staff to make clinical decisions Each tablet is disinfected with alcohol after every use, reducing the risk of cross-infection and enhancing the efficiency of the quarantine station The hospital's Wuhan pneumonia screening shows a sensitivity and accuracy of up to 80 and 90, respectively The 'Chest X-Ray AI Interpretation for Pneumonia System Model' developed by the hospital's Department of Radiology, with active participation from Professor Yong-Nian Sun's team at the College of Electrical Engineering and Computer Science Utilizing a tuberculosis X-ray AI auto-interpretation model developed by a previous AI biotech medical innovation research center project, it was adapted to the hospitals' pneumonia imaging data The collaboration between the parties ensured rapid completion Currently assisting in over 152 suspected Wuhan pneumonia screenings, sensitivities and accuracies of up to 80 and 90 have been achieved, respectively Moreover, for students conducting in-home quarantine at school dormitories, the university has adopted a smart monitoring approach with a 'Warm Heart Smart Bracelet' developed by a cross-disciplinary team, which continuously monitors quarantined individuals' body temperatures and heart rates as indicators for predicting symptoms When a rise in body temperature is detected, individuals can proactively confirm abnormal symptoms via a smartphone app and be prompted to seek medical attention Currently, bracelets are collected weekly and data is centrally uploaded to a cloud platform by the management staff for ongoing tracking, wholly enhancing the level of pandemic control internally and externally The university's cross-disciplinary team uses 'Warm Heart Smart Bracelets' to implement home quarantine policies effectively Photo source Official website「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

"Good technologies and applications must be developed outward to serve society" The initial group of people followed the professor's footsteps in the laboratory to study popular contemporary technologies and applications Shu-Fan Lin, CEO of ZeroDimension Tech Co, Ltd, believes that the implementation of AI in work reporting not only helps companies achieve an unmanned factory, but goes a step further to achieve a worry-free factory AI IoT tradition to the future Learning AI technology with Professor Chou at Tunghai University and often going to factories for observation and internships made Shu-Fan Lin realize "Why not try to solve it using AI or IoT" He noticed the pain points on site and hoped to use new methods and new technologies to solve existing problems, working with customers to find different solutions that are consistent with on-site operating models and improve past defects This is the origin of ZeroDimension Tech In the past, on the production line, data basically relied on manual writing, but now using machine vision recognition, data that could only be copied with pen and paper in the past can now be recorded first-hand on-site through lenses and mobile devices and sent back to the situation room in real time The status and problems of various indicators, such as efficiency, utilization rate, and OEE, in each factory are reported in real-time, forming a highly integrated system AI work reporting system improves data collection efficiency by 10 times Shu-Fan Lin said that the recovery and processing of on-site production information is a major pain point due to oil contamination on paper and the difficulty of recognizing handwriting on reports filled out by on-site workers At present, ZeroDimension Tech's AI work reporting system is being introduced into mechanical forging plants, using visual recognition methods to identify counters and workpieces and report work, improving the efficiency of information collection In addition, manual filling of reports has been converted to work reporting on tablet PCs, which greatly reduces the difficulty of information collection The collection of information in the traditional 3K dangerous, hard, dirty factory environment is a big problem From the collection and integration of production information to the generation of the final report, it will take one hour at the fastest or as slow as two to three days using manpower The development of digitization and AI can increase the frequency of information collection to once a minute Depending on the amount of information, reports can be generated in about 5 minutes and provide the latest information, improving efficiency by about 10 times AI work reporting system In the current stage, the AI work reporting system can be applied to aspects of factory production and on-site, and AI can be applied to counter recognition, work piece recognition, and facial recognition for work reporting At present, more AI applications and AI work reporting systems are being integrated to create a friendly and intelligent user experience, thereby reducing the workload of on-site operators and gradually making the site paperless Clear information collection improves the speed of decision-making by senior managers in the situation room ZeroDimension Techrsquos future development plan is to simultaneously improve the production capacity and efficiency of companies As for the difficulties and challenges encountered during the development process, Shu-Fan Lin pointed out that although machine vision is able to record and store on-site images, the large amount of dust in the environment on site can easily cause inconvenience in photographing images The dust and stains had to be removed from time to time to store clear images for recognition It is hard to overcome technical problems of the environment and is a goal that the company must towards in the future

Video surveillance products generally only provide a black screen for comparison Whether an image has mosaic or snowflakes can be determined by using the xception model for image quality monitoring, and the application of AI for determination allows the product to stand out Implementation of smart identification to monitor streaming in real time Point Media Technologies is a company that specializes in the development and manufacturing of video streaming products, and has achieved good results in the radio, television, and OTT fields The start-up coincided with the transition from analogue to digital television in Taiwan in 2013 Taiwanrsquos radio and television industry has always used foreign products, and not many products were developed by Taiwan This market gap led to the establishment of Point Media Technologies Video surveillance products generally only provide a black screen, video without audio, and audio without video, providing real-time images for determining mosaic Whether an image has mosaic or snowflakes can be determined by using the xception model for image quality monitoring, and the application of AI for determination allows the product to stand out Illustration of viewing audiovisual products Image quality monitoring can be used in audiovisual transmission to assist in quality monitoring At present, channel transmission operators use three methods are used for audiovisual transmission satellite, data line and Internet public network transmission The audiovisual transmission process goes through many encoding and decoding devices The operation process of these devices may cause damage to the image In the past, poor signal quality was only discovered after outputting images to the user end The inspection and repair process also takes a lot of time, resulting in poor viewing performance Due to the low cost of using AI for determining image quality, it can monitor each audiovisual node Once the system detects an error signal, it can notify the engineer immediately to deal with it, reducing the processing time and is better able to improve user satisfaction with the viewing experience High-quality audiovisual effects at only one-tenth the cost To achieve the high image quality required by radio and television, it often costs up to NT1 million to purchase related equipment After the introduction of AI technology, the cost is only NT100,000, which is only about one-tenth of the cost The AI audiovisual determination module has a multi-screen monitoring system After commercial verification, this AI image quality determination module is able to assist automated monitoring and improve the quality of audiovisual transmission If this AI module program is applied to a microcomputer, it will be more convenient to introduce it to various user units All radio and television, OTT, and live streaming operators can use this system to meet their automated monitoring needs The AI module can detect image problems with an accuracy of about 96, allowing problems to be quickly detected and resolved System architecture chart Channel transmission operators are responsible for the reception and transmission of dozens of channel signals in Taiwan The audiovisual signals received will be transmitted to various cable TV and OTT platforms in Taiwan Since each channel needs to be transmitted to many nodes, the TV wall in the monitoring center is full of TV signals Manual monitoring is imperfect, resulting in line abnormalities and unstable signal quality, which in turn affects the viewersrsquo rights After introducing the system, the image quality monitoring system replaces manual monitoring and immediately reports any abnormalities in the image, greatly improving the quality of audiovisual transmission Dual mode error event examination Point Media Technologies stated that the system is currently only able to monitor defects, snowflakes, mosaic noise, gap compensation, and jitter It is not yet able to automatically adjust quality, which will be the greatest challenge for future monitoring systems

As the COVID-19 pandemic continues to spread globally, confirmed cases and deaths are steadily climbing, causing widespread alarm The Taiwanese company Aivetech accidentally became a market sensation with its 'BRISE AI Anti-Allergy Angel' air purifier, originally launched to combat influenza and enterovirus, achieving rapid sales and going out of stock within less than three months, sweeping up 3-4,000 units off the market 'We didn't expect the epidemic to escalate so rapidly we initially just hoped to alleviate the difficulties of allergic families in Taiwan, and also purify influenza and enterovirus We wanted the BRISE AI Anti-Allergy Angel to be a guardian of family health, but the spread of the coronavirus unexpectedly boosted our popularity,' said Tsai Cheng-Chia, director of product and market development at Aivetech, in an urgent tone Aivetech, founded in 2013, transformed into a specialist in air detectors and air purifiers four years ago With many domestic and international brands in the market, Aivetech aimed to integrate artificial intelligence AI into its air purifiers, advancing towards becoming an AIoT Artificial Intelligence Internet of Things specialist manufacturer The purification rates of H1N1 and enterovirus reach 9999 Since early 2019, Aivetech has actively invested in the development of software and hardware for AI air purification In terms of hardware, the BRISE C360 can detect 7 types of air values PM 03 10 25 10 concentrations, harmful gases, temperaturehumidity, etc Most importantly, Aivetech spent half a year obtaining a test report from the P3 Laboratory-Guangzhou Institute of Microbiology that show high purification rates of 9999 for both the H1N1 influenza virus and EV71 enterovirus Regarding the currently prevalent new virus strain, Tsai Cheng-Chia mentioned that the strain has not been made public yet, so the labs are unable to test it However, for epidemic control, Aivetech is taking a dual approach on one hand, by connecting to the 'weekly nationwide clinic visits data' from the Disease Control Department to analyze big data during epidemic peaks, accelerating air purifiers during high transmission periods to filter out germs on the other hand, by enhancing the sterilization functions of special filters to help effectively control the epidemic BRISE AI Anti-Allergy Angel connects to open data from the Disease Control Department Tsai Cheng-Chia, product and market development director at Aivetech The BRISE AI Anti-Allergy Angel system has sensors installed both in the air detector and the air purifier, detecting indoor and outdoor temperature, humidity, and organic compounds, while also connecting to data from the Central Weather Bureau and the Environmental Protection Agency on climate, temperature, humidity, and PM25 air pollution Every 10 seconds, a piece of data is sent out, collecting various data and uploading it to the cloud system At the same time, using positioning systems, no matter in which district the air purifier and other devices are located, their temperature, humidity, and air quality can be clearly monitored Anti-Allergy Angel system fully protects the health of children Aivetech stores the collected data in the system and prompts users through the mobile APP, asking about the temperature, humidity, and PM25 levels that trigger allergic symptoms like sneezing, runny nose, and itchy eyes in family members with allergies, such as allergic children By using AI deep learning, it fully captures the environmental factors that induce allergic responses, whether due to a significant indoor-outdoor temperature difference or poor air quality By collaborating with nearby community clinics, including otolaryngology, family medicine, and pediatric allergy immunology specialists, data is provided to the community clinics, enabling parents and doctors to understand allergens and protect children's health holistically Currently, 8,000 allergy-affected families in Taiwan have adopted the BRISE Anti-Allergy Angel, and over 300 medical institutions use BRISE air purifiers BRISE AI Anti-Allergy Angel system Tsai Cheng-Chia states that BRISE is the world's first AI-driven air purifier, unlike regular air purifiers, it automatically learns the habits of its users, providing the most personalized clean air service, integrated with real-time chat services on mobile apps for more accurate anti-allergy and epidemic prevention services While air purifiers can block or kill airborne microorganisms and germs, they are ineffective against viruses and bacteria that land on surfaces Aivetech plans to launch a sterilization and deodorization machine in the second quarter of 2020, using short-wavelength ultraviolet light combined with a special catalyst formula to break down air molecules, reducing the activity of hydroxyl radicals and naturally achieving sterilization and deodorization The catalyst formula coating on the machine's equipment can last for 2-3 years without degradation Aivetech's team comprises professional physicians, environmental control experts, and AI technology specialists, collaboratively developing precise sensing and monitoring technologies for the home environment amidst the ongoing severe epidemic, continuing to contribute to the nation's health「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

Fast fashion in clothing, small quantities, diverse styles, short delivery times The textile industry faces the impact of the fast fashion trend among clothing brands, affecting the entire supply chain Global brand channels are promoting zero inventory, short delivery periods, and small-scale customization Balancing production time, quality, and cost is challenging Often, there is a discrepancy between ODM predictions and actual demands from brand owners, causing issues in material management and excessive inventory costs Due to inaccurate demand forecasts from customers, it often leads to difficulties in material preparation Excessive materials can increase leftover stock, while insufficient materials may delay delivery This project aims to establish an AI-based material demand forecast model specifically for major domestic manufacturers AI calculates sales trends to further predict demand The advisory team collaborates with Shentong Information Technology to mainly use the LSTM algorithm for the AI foundation The goal is to predict the next sales cycle based on past sales records, utilizing simple regression to complex 'Time Series Analysis' in statistics Usually, a period's sales volume closely relates to the previous period's, unless there is a major event, in which case it would typically follow a pattern There are various patterns of sales volume forecasts, including revenue, profit, customer counts, park visits, sales numberamount, etc This will take the example of a factory's monthly shipment batches, using the LSTM model to predict the next month's shipment batches Material Demand Analysis Execution Framework This project plans to establish a customer-specific material demand AI prediction model During the planning phase, three different machine learning algorithms were used to prototype the AI model Logistic Regression Algorithm Gradient Boosting Algorithm Deep Learning Algorithm Material Demand AI Prediction Model Planning Demand forecast error reduced from a maximum of 70 to 35, significantly reducing inventory volumes This project estimates customer demands, required material types, supply sources, and customer delivery dates using machine learning to establish a primary material procurement prediction system It reduces the prediction error of demand from the top five international customers from a high of 70 to 35, significantly lessening the amount of inventory needed「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

Yunlin County Mailiao Fruits amp Vegetable Cooperation - Taiwan Lettuce Village, located in Yunlin Mailiao, is a legend of green gold on the banks of Zhuoshui River Every lettuce creates huge business opportunities for Mailiao Township with annual export value exceeding NT100 million By combining AI computing technology under the digital management model, Taiwan's first early warning system for lettuce pests and diseases was established to reduce the number of times pesticides are sprayed and greatly improve the quality of vegetables Lettuce garden by Zhuoshui River Mailiao Township is an extremely important place that produces vegetables for domestic sales and exports, and the main source of supply is Taiwan Lettuce Village, which covers an area of more than 300 akkers Under the efforts of farmers, the annual output is about 12,000 metric tons, in which 80 of is exported The remaining amount is sold domestically Lettuce has become an important export agricultural product of Taiwan Although the goal of Taiwan Lettuce Village is planned production according to purchase orders, there has been constant pressure from unstable production and quality due to unstable climate in the past few years Therefore, Taiwan Lettuce Village established the first lettuce pest and disease early warning system in Taiwan with the assistance of Info-Link Services Combined with smart equipment in the field and mobile monitoring software, field management can be carried out and warnings can be received at any time, allowing managers to quickly respond and reduce potential losses Green Miracles Cultivated by Intelligent Agriculture The smart agricultural cultivation management system developed by Info-Link Services has three major functions "achieving crop yield prediction," "real-time warning of crop diseases," and "accurate smart prediction" Mobile phone field reporting AIHUB01The database of Taiwan Lettuce Village has a total of 28,702 cultivation records, and has produced significant benefits It not only increased manpower efficiency by 50, but also reduced the number of times pesticides was sprayed per period from 3 to 1 It also reduced costs such as fertilizer on average, costs were reduced by about NT10,000 per hectare At the same time, field work that was originally completely manual carried out was reduced to only requiring 20 manual operations, and 80 of the operations were replaced by information systems, greatly reducing manual operation costs Description of Benefits to Taiwan Lettuce Village AIHUB02Info-Link Services has completed the database of Mailiao Fruits amp Vegetable Cooperation, and also completed the crop yield prediction and disease prevention and early warning system, allowing vegetable farmers to make inquiries and reports on-site with their mobile phones Decisions on harvesting and prevention measures are made based on the data provided Early warnings for the prevention of disease can also significantly improve the use of fertilizer and pesticides by farmers solely based on feeling The lower frequency of pesticide use improves the overall quality of the ecological environment, and at the same time ensures the food safety of consumers

Labor-intensive, prone to oversights and errors, low shipping efficiency A domestic hook-and-loop tape traditional manufacturing transformation and brand management company has expanded new markets and business opportunities through the e-commerce platform model This requires reliance on substantial labor for product listing, order organizing, inventory management, and shipment tracking This results in limited product varieties and quantities that can be handled, and manual data entry is often prone to oversights or errors, affecting shipping efficiency and customer satisfaction, which is critical for the competitive advantage of the business in e-commerce Internally, many operations rely heavily on repetitive tasks across various computer systems, web pages, emails, etc Currently listed on 15 e-commerce platforms, updating single e-commerce information alone requires 2-3 months over 200 items, making rapid expansion difficult limited by manpower, product information is not detailed enough, leading to doubts in e-commerce reviews, affecting orders and subsequent satisfaction Presently, orders are only confirmed once a day, leading to an information gap of up to 24 hours Annually, there are over ten thousand orders to process into shipment orders, typically accumulating for 15-30 days before once grouping deductions from inventory, resulting in always inaccurate stock levels Streamlined Client Interface, Accelerating Implementation Efficiency The mentoring team collaborates with Ruijing Engineering Technology to integrate AI and RPA technologies through a web-based architecture Robotic Process Automation RPA applications are not installed on the local desktop but are stored on a server and accessed only when needed by the user This technology, known as Thin Client, provides higher performance and security compared to the Thick Client, which requires downloading applications and data to the local desktop The Thin Client does not require downloads on the local machine RPA collaborative service features include Web Scraping Complex web data collection and arrangement Email manipulation Data analysis and disassembly of content and attachments Web operation Precise and rapid web operations or filling in specific fields Application operation Timed positioning operations of other window applications Data processing Data format conversion, decomposition, and reassembly File Exchange Management Timed file production, adddeletemodify, FTP uploaddownload Database operation Heterogeneous database data exchange, read or write to a specific DB Data recognition Fixed format field data processing screenshot, snapshot, alphanumeric text parsing and recognition Scheduling Can be timed, repeated, cross-process all the above processes Alert mechanism Email, Line Notification etc designated or broadcast notification Software Robot Technology Solution Execution Architecture AI software robots enhance the processing speed of orders, inventory management, and purchasing in manufacturing operations, developing automated services to avoid data duplication and input errors, and seamlessly integrating processes across systems operating 247 The war room panel facilitates statistical analysis and real-time sales conditions on each e-commerce platform, predicting and optimizing product inventory Direct Purchase Order Process Automation Robot E-commerce Information War Room Statistical Analysis Dashboard Software Zero Errors, Reducing Costs by 15 to 90 面對快速變化又競爭激烈的市場環境，更需要減少重複性、低產值的工作，將人力運用在更高價值的工作上。 Facing a rapidly changing and highly competitive market environment, it is essential to reduce repetitive, low-value tasks, focusing manpower on higher-value work RPA software robots are 15 times more efficient than indirect staff, also enhancing process quality to near-zero error rate execution quality, offering opportunities to reduce costs by 15 to 90 Since it doesn't require significant changes to existing workflows, businesses generally do not need to spend substantial manpower on retraining or adapting to new workflows, which contributes to a higher acceptance rate among businesses Even in software deployment, it only takes about 4-5 weeks to go live「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」

Miniaturization of products, client demands full inspection A listed electronic component manufacturer in Taichung, responding to the 5G era injecting new growth momentum into the quartz component industry, especially under the explosion of 5G opportunities, the importance of quartz components will play a more crucial role than in the past in consumer products As frequency components move towards miniaturization and at the same time demand high precision, the manufacturing processes are more susceptible to subtle factors, necessitating manufacturers to manage comprehensive data across all aspects including human, machine, material, method, and environment to quickly identify key defective factors in complex production environments Differing perceptions of defects, difficulty in enhancing quality consistency With the trend of miniaturization and complexity of electronic components, visual inspection on the production line has four main functions including measurement, identification, positioning, and inspection, with inspection being the most challenging part as most electronics manufacturers still rely on traditional manual visual inspection Taking the PCB industry, where Automated Optical Inspection AOI technology has the highest penetration rate, as an example, a research institution once investigated and found that when two individuals inspect the same PCBA board four times, their mutual agreement rate was less than 28, and the self-agreement rate was only about 44 Due to differing perceptions of defects among on-site personnel, even automated machine vision can still lead to inconsistencies in product quality due to system settings or differences among quality control staff 偲捷科技檢測AI化，降低過篩率2030 With the support of the advisory team, collaboration with Sijie Technology aimed at the defects in the sealing process Based on CNN Convolutional Neural Network, the integration of multiple models introduced an AI recognition module to aid in the optimization of subsequent AOI tests, aiming to improve the accuracy of inspection equipment It is estimated that after introducing AI visual recognition, the over-screening rate could be effectively reduced to 2030 Thus, the industry, needing smarter inspection systems, has started applying AI technology to assist AOI equipment in optimizing subsequent screening tests AI-powered AOI Inspection Solution Cross-Model Design Concept Sealing AOI Inspection Trial Results Reducing false rejects, cutting manual screening workload by 50 The project, through a deep learning network architecture, reclassifies defects detected including true and false defects, and further classifies them to reduce the false reject rate of the traditional AOI solution This is anticipated to further aid manual inspectors in reducing more than 50 of the inspection screening volume, addressing current production line issues of relying heavily on manual re-inspection and low efficiency Future goals include integrating robotic arms for automatic loading and unloading, and analyzing defect causes to optimize production process parameters「Translated content is generated by ChatGPT and is for reference only Translation date：2024-05-19」