TECH OFFERS

Changes in consumer behavior and surging industrial demand are driving forces in advancing biomanufacturing technological advances. Enzyme-based biocatalysis is the centerpiece of many such industrial applications. However, while improvement in quality attributes could be observed by enzymatic biocatalysis, natural enzymes are often susceptible to low catalytic efficiency due to externalities ranging from temperature changes and pH levels. Despite recent breakthroughs in enzyme engineering, the conquest to identify the right enzyme could still be a stumbling block for many firms. Screening millions of variants to discover the right enzyme requires years of R&D, potentially setting the company back by a few million dollars. A startup has developed an enzyme engineering platform to identify and engineer novel enzymes to counteract these issues with an emphasis on their environmentally friendly process. With proprietary microfluidics technology, the enzyme engineering platform can build and test millions of enzymes ten times faster than existing technologies while boosting the chance of success in developing the most effective enzymes by 200 times. The startup supports the development process ranging from the screening of enzymes variants to full enzyme engineering, and scaling of desired enzymes to accelerate sustainable biomanufacturing for a wide range of applications. The company is seeking R&D collaborations with partners seeking the forementioned developmental expertise.

Virtual Reality/Mixed Reality (VR/MR) solutions involve high setup costs and a long development time; typically requiring between 4 weeks to 6 months just to construct a training scenario which is frequently prohibitively limited in scope and flexibility. With such long lead times, the application of scalable medical training is a challenge for the medical training industry - limiting access for many medical practitioners. This technology offer is an AI-powered medical training simulation utilising mixed reality and virtual reality to improve healthcare training for healthcare workers. It enables the creation of a wider variety of medical scenarios via its virtual patient engine and understands the verbal responses of medical learners via a conversational AI engine which also recognises a dictionary of medical phrases and drug names that are relevant to a clinical summary.

An AI-based system using acoustic pattern recognition has been developed to enable Computer Numerical Control (CNC) machine to possess self-diagnosis and provide feedback for machine operation. CNC machines are widely used in the manufacturing sector and they are machine tools that cut or move material as programmed on the numerical controller. Sounds associated with various cutting conditions such as normal cutting, heavy cutting, tool wear, and tool crash, can be recognised by the system, and this information can then be feedback to the actuators or end users for necessary actions. The system would facilitate remote monitoring of the machines, predictive maintenance, and failure analysis of the machine tools. System could be further adapted or enhanced to work for different types of instruments and acoustic sounds.

A new approach to carry out high throughput screening for COVID-19 pneumonia in large numbers of suspected COVID-19 patients is developed that will quickly triage patients with Chest X-ray or CXR radiographic findings (prior to a RT-PCR diagnosis) to stratify for high-risk patients and thus optimize hospital resources. The technology is developed on carefully curated Covid-19 pneumonia and other chest X-Ray images, by AI technology from deep transfer learning to seamlessly provide the radiology feedback for radiologists and clinicians. The software system is currently deployed in clinical environment, high performance achieved at Appropriate Use Criteria or AUC > 0.94. Industrial partners are sought to commercialise this technology to bring benefits to the healthcare industry.

Current methods of monitoring construction safety and productivity are tedious, costly and prone to human errors. Resulting in operations being non-compliance, dangerous and inefficient that leads to project delays, cost overruns and even reputational damage. This technology enhances operational safety and productivity for construction sites by leveraging on the video feed of existing CCTV on-site. It is able to conduct video analytics on the CCTV feeds in real-time and provides immediate actionable insights such as alerts, trends and reports. Site personnel and managers leverage this technology to evaluate sub-contractor performance, monitor vehicle movement within and around site, track construction progress, educate workers on unsafe practices and strengthen their safety policies.

One of the main challenges for space exploration, navigation and communication is to have a precise and reliable timing reference. An error of even one second can cause a huge difference for satellite signal transmission and navigation as satellites are moving at extremely high velocity. With the growing popularity of nanosatellites (1 to 10kg), the demand of reliable and precise miniaturized timing reference for nanosatellites applications increases exponentially. Deployment of traditional atomic clock onboard nanosatellites is certainly not possible due to the limited space, weight and power budget. This technology offers a space-hardened, small size, low power and highly precise version of timing reference module based on chip-scale atomic clock (CSAC) technology. The module design could be customised according to application requirements. The clocking performance of the developed timing reference module was thoroughly tested in lab and was successfully demonstrated in Low-Earth Orbit at altitude of 400km onboard the SPATIUM-I CubeSat. This technology can also be laterally applied to other harsh environment applications like mining, deep ocean application etc. Technology owner is seeking collaboration with company that are interested in incorporating this technology into their products and/or systems for accurate and reliable timing reference.

Globally, food security presents a perennial problem - one that can be addressed by improving crop yields. Yet in recent years, improvements to crop yield have plateaued and this is primarily attributable to plant diseases. Plant disease remains a challenge to detect simply because most inspections are done by the human eye - at the point of detection, plants are already in advanced stages of the disease, and even the most experienced food producers can often miss subtle, early-stage signs of plant infection. This technology offer is an automated imaging system combining cameras and three-dimensional (3D) multispectral imaging techniques that are far more sensitive than the human eye, enabling laboratory-level analysis of plant health in the field. The technique works because of the way light is reflected under different wavelengths of light; providing plant health information such as nutrients and diseases at a cellular scale. Unique identifiers of specific plant diseases, as well as plant stress factors, are identified by machine vision before they are even detectable by the human eye. Such early identification of plant health issues is crucial to enable corrective action to be taken at the optimum time.

Fouling is the accumulation of undesired materials on solid surfaces. In powder processing, fouling is a serious and common issue. This is caused by powder adhesion on processing surfaces, leading to clogging and disruption. Most of the time, powder adhesion is due to triboelectric charging. Electrostatically charged powder adheres to surfaces, accumulating and eventually leading to blockage of channels (e.g., transfer pipes, etc.). When fouling occurs, companies are forced to pause production lines for cleaning and unclogging which leads to productivity losses, energy costs, and manpower expenses. In addition, electrostatic fouling issues have been reported from electronic, plastic and pharmaceutical manufacturing industries as well. This technology is a customisable polymer coating that reduces the amount of static charge generated when a material contacts the coated surface. By reducing the formation of electrostatic charge, adhesion of materials is reduced, and thus the fouling issue is mitigated as well. The technology provider has currently developed prototypes that have helped solve static charge related issues in pharmaceutical, electronics, food and biomedical research industries respectively.

Membrane distillation (MD) is a membrane technology based on the vapour pressure gradient across the porous hydrophobic membrane. MD offers several advantages such as lower operating pressures and insensitive to feed concentration for seawater desalination. However, the commercialization of MD process has been constrained mainly by the lack of commercially available high-performance MD membranes and high energy consumption. Current state-of-the-art lab-scale fabrication of superhydrophobic membranes for membrane distillation often involve complex surface modifications and/or the massive usage of nanomaterials. However, these methods are often difficult to be scaled up. Hence, a pure rheological spray-assisted non-solvent induced phase separation (SANIPS) approach has been developed to fabricate superhydrophobic polyvinylidene fluoride (PVDF) membranes. The resulting membranes are found to have high porosity, superhydrophobic, high liquid entry pressure and hierarchical micro/nanostructures and can be easily scaled up. This facile fabrication method is envisioned to pave the way for large-scale production of superhydrophobic membranes for membrane distillation.