TECHINNOVATION TECH OFFERS

The DeepGI system is a cutting-edge AI tool designed to help detecting images captured during endoscopic examinations. Utilizing deep learning models, it can detect abnormalities in both the colon and stomach with over 90% accuracy. This real- time alert system assists medical professionals in identifying polyps with the potential to develop into cancer (neoplastic) and those without such risk (hyperplastic) in the colon. Additionally, it can pinpoint areas in the stomach where precancerous conditions, known as Gastric Intestinal Metaplasia (GIM), may be present. DeepGI is a vendor-unlocked system, compatible with most endoscopic cameras.

Oxygen enrichment membrane technology is emerging as a promising, cost-effective, and energy-efficient method for producing oxygen-enriched gas (OEG) with oxygen purities of 30-45%. Traditional oxygen production methods, such as cryogenic distillation and pressure swing adsorption, are often costly, energy-intensive, and complex, making them less suitable for applications requiring moderate oxygen enrichment. This innovative technology addresses these challenges through a thin-film composite (TFC) hollow fiber membrane that incorporates a novel use of polydimethylsiloxane (PDMS) as a selective layer on a polyethersulfone (PES) substrate. The PDMS selective layer is applied using a flow coating technique, which is both simple and scalable, allowing for consistent production of high-performance membranes. The technology was upscaled to commercial-sized membrane modules producing 15-53 Nm³/h of OEG with oxygen purities between 31-38%. The membrane system operates at ambient temperatures and pressures, offering significant energy savings and reduced operational costs compared to traditional methods. The benefits of this technology are substantial, including improved cost-effectiveness, enhanced energy efficiency, and flexibility in scalability, making it suitable for a wide range of industrial applications.  The technology owner is seeking collaboration with membrane manufacturers to further scale up this innovative technology, and with end-users who have a demand for oxygen-enriched gas with 30-40% O₂ purity.

Faced with the increasing levels of carbon dioxide, carbon capture, utilisation, and storage (CCUS) technologies have garnered significant attention. However, as most CCUS technologies rely heavily on various forms of monetary support from governments and faced numerous technical and scalability challenges, most of the CCUS facilities developed are unable to achieve financial profitability or even achieve a net reduction of carbon dioxide (CO2) emissions. The technology proposed herein relates to an electrochemical-based CO2 reduction reaction process, which can directly capture and convert CO2 to carbon nanotubes (CNTs), a high-value material that exhibits unique electrical and thermal properties suited for applications in various sectors, including electronics, energy storage, sensors and medical uses. In contrast to synthesis methods that involve complex reactions and expensive catalysts, the proposed method uses a molten salt chemistry that can convert CO2 to cathodic solid carbon nanotubes (CNTs) via the electrochemical process. Although high reaction temperature (about 760 degC) is required, this method is highly controllable and uses cost-effective pure iron catalyst, producing high quality CNTs at a relatively high production rate. Based on preliminary process modeling and technoeconomic analysis, this technology has the potential to be completely CO2-negative without re-emission, is more scalable, and profitable with high quality CNT materials. The technology owner is seeking to collaborate with industry partners and research institutions for joint R&D to advance the lab scale technology to pilot or event industrial production scale, as well as to explore applications for the CNTs produced. Upon further development, the system has the potential to be integrated with existing carbon capture systems to improve their financial viability and achieve carbon negative objective.

Acknowledging the importance of high-quality data, this project aims to revolutionize data lifecycle management in the AI to improve data accessibility, collaboration, and commercialization. The solution enables (i) efficiently clean, process and extract valuable data assets from high volumes of mass data, and (ii) contribute and commercialize high-quality data assets without disclosing the actual data. DataS comprises three pillars: (1) GLASSDB serves as an end-user database, including add-in tools for data cleaning, visualization, security, aiding data owners in preparing data for future transactions. (2) Apache SINGA offers a powerful machine learning library to allow users to efficiently apply or develop AI models on their data. (3) Falcon enables privacy-preserving federated learning. It allows multiple parties to develop AI applications using joint data without compromising privacy.

Monitoring for product or part surface defects and anomalies such as cracks and chips throughout the manufacturing process is vital for product quality assurance and control. Traditional inspection or machine vision systems often struggle with complex and nonlinear defect patterns, leading to false positives and missed defects. Deep Learning AI-based detection methods, particularly those using deep neural networks (DNNs), typically require a sufficiently large amount of labelled training data to be effective. However, gathering and labelling such data can be time-consuming and costly, especially for rare or specialized defects. The technology owner has developed a cost-effective system solution leveraging on neuromorphic AI to utilise the principles of human cognitive memory with machine learning to detect surface defects and classifies them reliably and accurately. The system solution includes their patent-pending neuromorphic AI framework architecture with complementary hardware modules, patented lighting system and proprietary software platform. Through the use of neuromorphic edge-AI chip, the proprietary AI model requires smaller training dataset supported incremental learning capabilities, resulting in a high precision, high accuracy system overtime. As the system is camera agnostic and customisable, it enables easy integration and retrofitting to various industrial applications. There are currently a few ongoing POC projects with industrial partners for automating and enhancing quality checks within their manufacturing line. They are seeking industrial collaboration opportunities who are open to explore surface defect detection for quality assurance or monitoring applications.

The demand for industrial automation and robotics has been steady increasing across various industries to automate manual work. By embracing autonomous robots, like Automatic Guided Vehicle (AGV) and Autonomous Mobile Robot (AMR), companies can reduce labour cost and enhance safety while increasing operational efficiency. However, this system can be costly, complex to integrate and scale into existing operations and require technical know-how for their operation. The technology owner has a technology solution focusing on AGV and AMR for industrial automation that aims to address existing adoption challenges. The technology solution provides a plug-and-play approach which comprises of a customisable AI-powered robot controller which can be used to build new industrial robots or to retrofit into existing deployed ones (e.g. AGV). This approach is cost-effective and simplifies integration into current operations. Using a Simultaneous Localisation and Mapping (SLAM) navigation system with ready-to-use software assessable via a mobile device, the technology solution is user-friendly and allows less experienced personnel to operate it easily. Their AI capabilities also enable autonomous decision-making while utilising data analytics for operational efficiency. The technology owner is looking for collaboration partners, such as industrial system integrators, who are keen to explore customisation of AGV and AMR solutions aimed to improve operation efficiency and reduce cost.

Force sensing is widely used across various applications. In recent years, the demand for automation and labour-saving solutions has rapidly surged, driving the growth of markets for human-like robotic hand replacements. Additionally, extended reality (XR) and gaming controllers are striving to enhance the immersion through pressure sensing. As a result, there is an increasing need for force sensing technologies that closely mimic human sensation. The primary methods for force detection rely on capacitive and resistive sensors. These sensors measure force by detecting changes in capacitance or resistance values. However, they face challenges like low surface resolution and nonlinear responses to varying load ranges, leading to a perception that differs from natural human touch / operation. To address these challenges, the technology owner has developed an advanced capacitive force-sensing technology that utilizes micro-pillars (micro-protrusions) which are just tens of microns in height. This technology accurately detects changes in capacitance at low load ranges by leveraging micro-pillars. These micron-scale structures are formed using conductive rubber through an original micro-molding process. When configured in a grid layout, the technology ensures high linearity and surface resolution down to a single digital pitch, enabling force sensing that closely replicates human perception. The technology owner is keen to collaborate with industrial partners across various sectors, such as ICT manufacturers, robotics companies, and manufacturers of controller and electronic instruments, to explore potential applications.

The Edge AI-based Drone System for Pipe Inspection and Monitoring addresses the need for efficient, accurate, and real-time pipeline infrastructure inspections. It leverages edge AI processing to detect defects, offering a significant advantage over traditional methods. Unlike conventional systems that rely solely on optical cameras, this solution integrates both optical and thermal imaging, enhancing the detection of various pipeline-related issues. The system’s unique value proposition lies in its ability to process data locally on the drone, ensuring immediate issue detection and minimizing data breach risks by reducing the reliance on cloud processing. The technology owner is seeking collaboration with SMEs specializing in drone manufacturing, AI and machine learning, thermal imaging, industrial inspection services, telecommunications and IoT, and data security, which offer complementary expertise for the development and commercialization of the technology.

A variety of force sensors have been adopted for different applications. Among them, sheet-type sensors, designed to be installed on curved surfaces, are typically categorized into two types: resistive type sensors that rely on the deformation of conductive materials, and capacitive type sensors that utilise the deformation of elastic materials. However, both types face limitations due to the restricted amount of deformation of the materials used. Consequently, these sensors often have a narrow force detection range (dynamic range), making it challenge to measure both low and high forces with a single device. To address this, the technology owner has developed an innovative, large-area force sensing device leveraging a unique built-in wire structure. This device is capable of detecting surface pressure distribution in a matrix with high linearity across a broad range of loads, from low to high. The detection level of capacitance exhibits excellent linearity against applied loads. By integrating conductive wires into a large elastic sheet, the device is scalable, making it suitable for large-area applications. This technology offers versatile applications across various sectors. In smart retail shelves, it can monitor product status and inventory levels over the entire surface. In robotics, it enables full-body tactile sensing for the entire surface of a humanoid. In automotive applications, it can detect and monitor the status of drivers and passengers. The technology owner is seeking R&D collaboration and licensing opportunities with industrial partners, particularly in robotics, automative, retail and logistics sectors, to explore potential applications.