TECHINNOVATION TECH OFFERS

In radiotherapy for patients with gastrointestinal (GI) cancer, real-time, continuous monitoring of X-ray radiation in the GI track can greatly improve the precision of the treatment. This proposed technology consists of a swallowable X-ray dosimeter capsule for real-time monitoring of absolute absorbed radiation dose and changes in pH and temperature in the GI tract. Using a neural network-based regression model and a luminescence of nanoscintillators fiber, the capsule is able to estimate radiation dose from radioluminescence and afterglow intensity and temperature. Initial preclinical study in a rabbit model showed that the dosimeter was approximately five times more accurate than standard methods for dose determination. Hence, these swallowable dosimeters may help to improve radiotherapy and understand how radiotherapy affects tumour pH and temperature. The technology owner is seeking for collaborations and out-licensing with medical institutions and medical device companies for clinical testing and further research identifying the capsule's position and posture after ingestion, developing a robust positioning system.

Thermal management is an essential part of the design of high power density electronics. As the power density of electronic devices increases, so does the amount of heat they generate, and this heat must be dissipated effectively to prevent the devices from overheating and failing. This technology offers a method to produce high thermal conductivity boron nitride (BN) composites that aim to improve thermal management in high power density electronics, leading to more efficient, more compact, and safer electronic systems. BN composites are a group of materials made by combining boron nitride with another material, such as a polymer, metal, or ceramic. A key advantage of such composites is that they exhibit higher thermal conductivity than any commercially available material that is electrically insulating. The resultant BN composites are also low in weight, easily shaped, exhibit good mechanical properties, and offer the unique capability of designing the path by which the heat will be conducted. These properties fulfil the demanding requirements for electronic packaging in emerging markets like Internet of Things and embedded systems, autonomous vehicles, high speed computers, satellites to name a few. The technology owner is seeking for co-development and out-licensing opportunities with semiconductor and device-assembling companies that require high thermal conductivity materials.

Lithium-ion (Li-ion) batteries are the most developed and widespread rechargeable batteries and is expected to dominate the market in coming years. Despite the wide adoption, Li-ion batteries face challenges that result in degradation of electrochemical performance due to side reaction with the electrolyte, dissolution of electrode components, transformation and pulverization of its structure and so on. The patent pending technology proposed herein aims at improving the performance of Li-ion batteries through the application of ultrathin oxide layers which increases the cycling stability and C-rate capability of lithium-ion batteries. The technology modifies the surface of the silicon-based anode of the Li-ion battery with a thin layer of Zinc Oxide (ZnO) applied by atomic layer deposition (ALD) technology. Typically, the ZnO thickness is of order of few nanometers. The ZnO layer suppress side reactions by limiting the excessive growth of the passivation layer at the anode interface and liquid electrolyte, inhibiting the dissolution of the electrode components, and enhancing conductivity and Li-ion transfer. This resulted in  increased battery capacity particularly during fast charging and discharging and increases cycle life of battery. The technology owner is seeking to license the technology to a battery manufacturing partner who have access to ALD equipment to facilitate the integration of the anode surface modification technology into their battery manufacturing process.

Recent advances in soft robotics revolutionize the way robots interact with the environment, empowering robots to undertake complex tasks using soft and compliant grippers. Compared to traditional rigid structures. Soft grippers have excellent adaptability for a variety of objects and tasks. However, the existing gripper systems faces some challenges, such as handling delicate, wet, and slippery items, the risk of damaging valuable items, and high production cost. Based on pneumatic jamming of 3D-printed fabrics, the technology owner has developed a variable-stiffness soft pneumatic gripper that can apply small forces for pinching and pick-up heavy objects via stiffening. The invented grippers are soft and adaptive to handle delicate items with various shapes and weights, minimising the damaging risk of items during the gripping process. In addition, such gripper with adjustable stiffness could handle heavy and bulky items by increasing its gripping strength. These benefits make the gripper more versatile and adaptable to various applications in agriculture, food processing, packaging, manufacturing, and human-robot interaction (HRI). The technology owner is seeking to do R&D collaboration, IP licensing, and test-bedding with industrial partners intending to integrate variable-stiffness gripper in their applications. 

A microfluidic chip-based mechanism has been developed as a Point-of-Care Testing (POCT) device to replace Lateral Flow Assays (LFA) for fast and convenient blood analysis. The microchip system utilises the principle of immunoassays but with high accuracy and compatibility to different signalling tags, providing a quantitative readout. Conventional immunoassays involve multistep procedure and long process time. While LFAs are fast and convenient, they are qualitative. The device demonstrated a one-step assay that can achieve equal or higher sensitivities than standard methods within significantly shorter total processing time. In a microfluidic device, the sample flows in precisely defined microchannels, which allow better control of fluid behaviour and higher consistency in testing results compared to LFA in which the sample flows by wicking through the porous paper-based material. This technology resides in the assembly of components and materials to immobilise antibodies or antigens onto the chip which can be easily scaled for commercial production. The technology owner is seeking collaborations with manufacturers of IVD devices or Medtech companies to out-license the technology and expand the range of antibodies targets for the microchip.

The technology consists in a new type of neural networks, providing explainable, verifiable, compact and private AI solutions. Explainability: the technology provides precise global explanations and the exact rules learned by the AI model, even with large datasets. We transform clients' raw data and/or models into meaningful results through high-quality visual analytics, empowering them to enhance the model based on these explanations. Formal Verification: the technology allows the client to formally verify certain properties of the model, such as its robustness to adversarial attacks, its fairness according to certain features, etc.

Neurological diseases are the second leading cause of death. CT scans have been used as the primary modality to diagnose brain abnormalities such as Intracranial Haemorrhage (ICH) and neurodegeneration. Radiologists usually have to deal with an overwhelming scan backlog and writing radiology reports is a time consuming process. Manual segmentation of lesions is tedious and existing heuristics have been shown to overestimate lesion volumes. Clinicians are also wary of the ‘black box’ nature of deep learning models. Hence, an automated tool in the workflow could substantially improve clinical productivity and interpretability is crucial to build trust with clinical stakeholders. Our proposed technology is an AI solution that automates ICH detection and brain tissue segmentation on CT scans, producing accurate volumetric information to assist triaging. Our technology also comes with a set of tools to interact with the AI models and generate reports easily. Moreover, we strengthen our AI transparency with interpretable models. Our platform also focuses on model robustness tests to assure AI safety.  

Digitalisation is a global trend with digital twin technology increasingly adopted in various industrial segments including smart factories and plants, digital facility management and operation & maintenance, building and construction, etc. Rapid generation of digital twin of physically existing is desired. Conventionally, digital twin is mainly generated using design software which requires professional modellers to spend substantial design time pending on the complexity of the physical twin (to be constructed) and the manpower available. Building information modelling (BIM) is increasingly used as a representation for the digital twin. For existing environments, scan to BIM technology and authoring software products are used for the process of reconstructing of BIM models from LiDAR scanned point clouds. This manual process is typically time consuming, tedious and error prone. Often, meshed models are used for visualization purpose of the digital twin.

Coronary artery disease (CAD) is the leading cause of death worldwide. Computed Tomography Coronary Angiography (CTCA), as a non-invasive alternative to invasive catheterized coronary angiography, has emerged as a recommended first-line investigation for CAD. However, the current practice of generating reports involves a time-intensive process, with CT specialists spending 3-6 hours annotating scans. Furthermore, there is a lack of effective tools for analysing coronary calcium scores, stenosis severity, and plaque characterization. This AI driven platform is for CT data processing that provides a streamlined 'one-stop' solution spanning from diagnosis to clinical management and prognosis. Its key features include: AI-driven platform for CTCA, catering to clinical, research, and industrial applications. Large, shareable, de-identified, Personal Data Protection Act-compliant real-world CT data. Precision toolkits for anonymization, coronary calcium scoring, epicardial adipose tissue (EAT), stenosis severity assessment, plaque quantification, CT fractional flow reserve (FFR), and reporting. The platform’s highly automated features assist physicians in interpreting and synthesizing large volumes of CT data, while minimizing bias, increasing reproducibility, and providing numerical insights in a graphical manner. It offers a comprehensive ‘one-stop’ solution for diagnosis and clinical management of CAD.