TECH OFFERS

One use of AI, including deep learning, is in prediction tasks, such as image scene understanding and medical image diagnosis. As deep learning models are complex, heatmaps are often used to help explain the AI’s prediction by highlighting pixels that were salient to the prediction. While existing heatmaps are effective on clean images, real-world images are frequently degraded or ‘biased’-such as camera blur or colour distortion under low light. Images may also be deliberately blurred for privacy reasons. As the level of image clarity decreases, the performance of the heatmaps decreases. These heatmap explanations of degraded images therefore deviate from both reality and user expectations.  This novel technology-Debiased-CAM-describes a method of training a convolutional neural network (CNN) to produce accurate and relatable heatmaps for degraded images. By pinpointing relevant targets on the images that align with user expectations, Debiased-CAMs increase transparency and user trust in the AI’s predictions. Debiased-CAMs are effective in helping users identify relevant targets even on images affected by different clarity levels and multiple issues such as camera blur, poor lighting conditions and colour distortion. The AI’s prediction also becomes more accurate. As the model is trained using self-supervised learning, no additional data is needed to train it.  The training for Debiased-CAM is generalisable, and thus applicable to other types of degraded or corrupted data and other prediction tasks such as image captioning and human activity recognition. Used to train a convolutional neural network (CNN) to produce accurate and relatable heatmaps for degraded images. By pinpointing relevant targets on the images that align with user expectations, Debiased-CAMs increase transparency and user trust in the AI’s predictions. It also increases the ability of meeting regulatory standards to deploy CNN models in the following applications, where explainable AI is required. Healthcare, eg. Radiology Autonomous Vehicles   Produces accurate, robust and interpretable heatmaps for degraded images Works on images with multiple degradation levels and types such as blurring and improper white balance Agnostic to degradation level, so that enhancement can be applied even when the level is unknown Perceived by users to be more truthful and helpful as compared to current heatmaps distorted due to image degradation Method of training can be applied to other degradation types and prediction tasks Explainable AI Infocomm, Video/Image Analysis & Computer Vision, Artificial Intelligence

Lists are an indispensable part of the online experience, often used to show many results, such as products, web pages, and food dishes. These items can be neatly sorted by a desired attribute like price, relevance, or healthiness. Listed items often have multiple attributes. However, instead of being able to sort multiple attributes simultaneously, consumers are currently limited to sorting only one attribute at a time. This makes searching for the desired item tedious and confusing. Imma Sort supports interpretable and multi-attribute sorting. Sorting for two or more attributes is possible. In contrast to existing search technology, Imma Sort trades off the smoothness of the sorted trend for the main attribute to increase ease of prediction for other attributes, by sorting them more approximately. Results for specific attributes can be made smoother by setting higher importance weights. Provides intuitively sorted results sorted by two or more attributes to improve decision-making and user experience Results can be customised by allocating higher weightage for selected attributes Enables users to perform multi-attribute sorting in any existing list interface without requiring sophisticated spreadsheets or data visualisations Can be integrated into search and recommendation systems across a wide range of applications Can also be incorporated into various search and recommendation systems for more effective search results. Examples of possible applications: Food dishes can be sorted by healthiness and tastiness Hotels can be sorted by price and distance Sorting by price and rating would generate results that generally trend in one direction for both attributes. This makes it easy for users to anticipate the values of multiple attributes as they move down the list, without having to construct a mental list for the secondary attribute. By decreasing users’ mental effort, this will improve decision-making and increase satisfaction. Multi-Attribute Sorting, e-commerce, algorithm Infocomm, eCommerce & ePayment, Enterprise & Productivity

Cadaveric bones are used to carry out medical training for surgeons and trainees. However, such bones are limited in supply, difficult to store, inconsistent in terms of quality and costly to use for repeated training. As such, it is necessary to create an alternative to cadaveric bones that is equally realistic while being more cost effective and easier to obtain. This technology can resolve the limitations of cadaveric bones by offering the formulation and processing method to produce a Fused Deposition Modelling (FDM) system-agnostic bone-like 3D printing filaments for surgical models printing. Printed anatomical bone models developed from this technology will have the look and feel of the real bone. The technology presents an affordable and readily available alternative that minimises the demand for cadaveric bones while still providing realistic training to medical professionals. The technology owner is seeking for collaborations with companies interested to scale-up the manufacture of the filaments and/or licensing of the technology.   The technology offers the formulation and processing method to produce bone-like filaments that allows users to 3D print realistic and relatively lower cost cadaver bone models for use in medical training of surgeons and trainees. These bone-like filaments have the following features: Compatible with all FDM 3D printers Achieves good mechanical properties  (~80% improvement in tensile strength) compared to the commercial bone-like filament. Printed bone models using this technology feel realistic to drill and cut and screws can be tightened nicely, just like real cadaveric bones. This technology is primarily targeted for use in the healthcare industry (hospitals, medical schools, and biomedical companies) to obtain realistic anatomical bone models for surgical training and workshops at a more affordable price compared to cadaveric bones. Surgeons can also create 3D printed bone structures to help them plan surgical procedures before operating on the patient to better understand the procedure. Significantly lowers the cost for realistic medical trainings and workshops as compared to using cadaveric bones Fabricated bone models are equal in quality with actual cadaveric bones cadeveric bones, cadever, cadevers, 3d printing, fdm, medical training, surgery, surgeon, medicine, bone models, additive manufacturing, fused deposition modelling, printed anatomical bone models, surgical models, printing, bone-like Materials, Composites, Manufacturing, Additive Manufacturing, Moulding, Sintering, Casting & Nanoimprinting, Life Sciences, Industrial Biotech Methods & Processes

This Technology Offer allows automation of the process of monitoring stock levels in the warehouse, and when they drop below a certain threshold, trigger a certain event.  This technology can be used to address the following challenges: What items are on the shelf? What is the current inventory level? What items need to be restocked? How many items need to be restocked? This technology can automatically trigger material top-up quantity from a cloud platform to suppliers, based on visual signals using webcams. Two types of technology concepts are used: 1) Webcam scans AR code to detect stock to be replenished or 2) Webcam compares area occupied vs empty (ROI image area mapping), and calculates stock to be replenished Thus, stock management can optimized and no longer require storage in warehouse; they can be directly sent to Centralized Packaging, hence saving staging space & inventory. This technology would be useful for companies that are interested in automating the warehouse processes. This in turn could reduce the time and manpower required to do inventory checks. The technology can also be applied in places with high turnover stock rates or unpredictable selling rates, e.g., pharmacy or supermarket. This technology allows for automatic monitoring and top up triggering of inventory levels, thus reducing time and manpower required to do inventory checks. Inventory Management, Logistics, Warehouse Management Logistics, Inventory Management

Human DNA contamination is an issue in the process of manufacturing devices for molecular and genomic applications. Such devices need to be certified human DNA free for sensitive downstream processes. Therefore, there is a need to monitor human DNA residual level in factory surfaces and in final products. The technology provider had identified a novel target for detection of human DNA and developed a highly sensitive PCR based detection method for presence of human DNA based on this target. They have demonstrated that the method is specific, sensitive and with low cost. Compared to a commercial kit available in the market, this method is at least 500 times more sensitive. This method will be a valuable tool for device manufacturers. In addition, it can also be applied on forensic as well as environmental samples. Furthermore, the principle of the method can be adapted to detect other animal species for either identification or monitoring purposes. The said method is simple and reproducible. It meets the requirement described in the ISO 18385 standard (Minimizing the risk of human DNA contamination in products used to collect, store and analyse biological material for forensic purposes-Requirement) in terms of detection sensitivity. One of the key advantages of this innovation is its superior sensitivity level in human DNA detection. When compared with a commercial human DNA detection kit on the market, thismethod demonstrated at least 500 times higher sensitivity when performed under comparable conditions in parallel. The primary application will be for certification of human DNA-free in manufactured products. With this test, products can meet higher standard of requirement especially in the molecular biology and genomic DNA related applications. Therefore manufacturers can broaden its market reach and add value to their products. This method can also be applied identification of human remains in forensic samples. With higher sensitivity level, efficient routine tests procedure can be established by using less samples volume or even pooled samples. The overall test cost will be lower.   DNA, DNA detection, PCR Healthcare, Diagnostics, Medical Devices, Chemicals, Agrochemicals, Analysis

Multiple Pickup and Delivery Planning (MPDP) is very common in both passenger and cargo transport operations. However, due to the planning complexity, it is impossible to quickly generate good solutions manually to support daily transport operation planning. The MPDP solution developed by the team is able to close the gap by quickly generating transport plans for multiple vehicles with multiple pickup locations and multiple delivery locations. Web-based application. No installation required and easy to access Capable of handling multiple complex constraints, such as different time windows, different vehicle types, different vehicle departure times, different vehicle start and end locations, Visualisation of jobs and planned routes on both map and Gantt chart Flexible for human planner to overwrite the existing plan with drag-and-drop feature The MPDP solution can be applied to two potential areas. The first area is passenger transport planning for Book-a-Ride, bus pooling and carpooling services. The MPDP solution is able to quickly generate high quality plans either in advance or in real-time to replace the manual planning. The second area is cargo delivery planning in the logistics industry. The potential users include packed food delivery to consumers from different restaurants, courier service providers who provide same day pickup and delivery services, premier transport service providers who provide 2-hour delivery services etc. The solution will facilitate the human planner to assign travel routes based on the vehicle’s current location, vehicle capacity, service requestor’s start and end location as well as service start time. The solution will also consider the drivers break time, vehicle departure location and end location. Moreover, the planning and scheduling engine developed is able to run in the background to assign passengers to vehicles automatically. This feature can also be used for the real-time planning of last minute transport service requests. The solution can be applied for both passenger service providers and logistics industry players for delivery planning. Potential users includes of Dial a Ride (DAR) service providers and logistics opterators that provide same day pickup and delivery courier services.  Provide better service to customers Reduce operating cost by cutting vehicle traveling distance on the road Improve vehicle utilisation to generate more revenue Infocomm, Enterprise & Productivity, Logistics, Planning & Order Processing, Transportation, Delivery & Distribution