TECH OFFERS

In the mid-infrared region, gases exhibit absorption spectral features that are typically two orders of magnitude stronger compared to the near-infrared region. This makes the mid-infrared quantum cascade laser (QCL) a highly suitable choice for gas spectroscopy applications. QCLs offer several advantages, including broadband spectral coverage ranging from 3 to 25μm, narrow linewidth, compact size, and robustness, which have contributed to their popularity in various spectroscopic applications. In this context, a portable gas sensor has been developed utilizing self-developed QCL arrays, covering two specific wavelength regimes: 9-10 μm and 13-14 μm. To further enhance the detection sensitivity, an artificial intelligence (AI) algorithm has been integrated into the gas sensor. The incorporation of a hollow-core fiber as a miniaturized gas cell contributes to the overall compactness of the system. By leveraging the capabilities of QCLs, this gas sensor overcomes critical weaknesses associated with existing approaches, particularly their lack of selectivity and inability to differentiate mixtures of gases effectively. We anticipate that this technological innovation will accelerate scientific research progress and prove valuable across various industry sectors. The innovation of this portable gas sensor is mainly in the laser source and beam combining approach. Compared with the commercial QCL products, the developed QCL arrays exhibited wide spectra tuning range, ultra-fast tuning speed, narrow linewidth, and eye-safe average power. To combine the laser beams in the array, a cost-efficient beam combining method has been developed. This method utilizes an aspherical lens and a series of mini mirrors to collimate the individual beams from the laser array. The system is controlled by a LabVIEW program, which simplifies its operation.  After conducting measurements, the AI algorithm automatically calculates the concentration of the target gases. This information is then displayed on the software interface, providing a convenient and user-friendly experience. The gas cell in the sensor employs a hollow-core fiber, which results in a quick analyte charging time of less than 1 minute. Furthermore, the gas sensor utilizes a broadband laser source, enabling simultaneous detection of multiple gases. The performance of the homemade QCL array is notable in terms of lasing peak and transverse mode, making it well-suited as the light source in gas spectroscopic systems. Notably, the gas sensor extends the operation wavelength regime into the ~13-14 μm region, which is advantageous for detecting volatile organic compounds (VOCs) that have strong absorption features in this range. In terms of detection limits, the gas sensor has been evaluated to achieve 940 parts per billion (ppb) for acetylene and 470 ppb for o-xylene. Primary application areas: scientific research, environmental monitoring, and industrial process control. Other areas: Indoor air quality monitoring and oil & gas. The potential products: Mid-infrared photoacoustic gas sensor, QCL-based dual-comb gas sensor, Cavity ring-down gas sensor and liquid sensors.   The global gas sensor market size was valued at USD 2.50 billion in 2021 and is expected to expand at a compound annual growth rate (CAGR) of 8.9% from 2022 to 2030. In this context, the QCL plays a pivotal role as one of the primary light sources in mid-infrared gas spectroscopy applications. Consequently, the QCL-based gas sensor has promising potential in the gas sensor market size. This technology is portable and provides both high selectivity and sensitivity with key benefit lies in three domains: Gas Sensing: this solution enables precise and accurate gas sensing, allowing for the detection and differentiation of multiple trace gases in various environments. Spectroscopy / Instrumentation: With the capability to design and create long-wavelength quantum cascade lasers, our technology is well-suited for advanced spectroscopy and instrumentation applications. IoT (Internet of Things) for Smart-Nation: By integrating this technology into the Internet of Things framework, contribute to building smarter and more efficient nations with improved environmental monitoring and management. The most critical problem of the existing technologies, such as electronic and chemical sensors, lies in their lack of selectivity. This means they are unable to distinguish between multiple trace gases unless more advanced methods like GC-MS or FTIR technology are employed. Unfortunately, these advanced methods are both bulky and expensive, restricting their usage to laboratory environments only. Quantum Cascade Laser (QCL), High sensitivity, Multi gases, Spectroscopy, Sensing system Electronics, Lasers, Optics & Photonics, Infocomm, Artificial Intelligence, Green Building, Indoor Environment Quality, Environment, Clean Air & Water, Sensor, Network, Monitoring & Quality Control Systems

Fast-moving consumer goods (FMCG) and other product components come in a wide variety of shapes, sizes and packaging configurations. During the manufacture of such products, a key challenge for automation is to effectively handle and manipulate such diverse products during production or logistical processes. Users planning to automate their production lines typically have to take into consideration the use of either multiple grippers for different product types, or incorporate an automated tool changer with added complexity and cost. To address this challenge, a Singapore start-up has developed a universal soft robotic gripper designed to manipulate a wider range of product sizes by incorporating a resizeable gripper base. Gripper adjustment is automatically carried out via an integrated computer vision system thus minimizing the need for human intervention during pick-and-place processes. The gripper's soft fingers also minimize damage to products during the gripping process. Vacuum Suction Gripper Incorporating extendable linkages in the gripper arms for resizeability, the gripping workspace remains adaptable to handle products of various sizes. The gripper arms may either take the form of fingers or suction cups configuration. With 5 vacuum cups embedded, the gripper is ideal to handle pouched products or carton boxes of various sizes. Gripper weight: 2.18kg Gripping width: 125mm to 315mm Manipulating weight: up to 10kg Actuation method: Clean, dry air up to 250kPa Operating temperature: up to 100°C Computer Vision System An integrated computer vision system provides the gripper with the ability to recognize the type, location, and orientation of the product to be picked, and commands the gripper to adjust the gripping space and pose to pick the product from the correct location. This process is fully automated without requiring human intervention. For increased utility, the computer vision system may also be configured to perform quality inspections of products being handled. The vacuum suction gripper can be used to palletize or depalletize carton boxes or pouched products (such as coffee powder, sugar packs, rice packs etc.) of various sizes up to 10kg. The computer vision system will be deployed if the working space is not in an organised condition (i.e., randomized locations and orientations of the products), such as when products are scattered in a tote bin, randomly positioned on a conveyor etc. The market value for FMCG robotic packing in the APAC market is estimated to be at USD 1.1 billion and the global market value is worth USD 7.8 billion. Sources: Cobots Transforming the Global Industrial Robotics Market—Opportunities Forecast (Frost & Sullivan) Passport, The Megabrands: The Top 100 FMCG Brands Worldwide (October, 2018) Technavio's library The reconfigurability of this gripper provides high adaptability to many applications, compared to conventional grippers with fixed gripper bases offering limited gripping ability for products of diverse shapes and sizes. Benchmarking tests have been conducted to compare the grippers with other commercially available grippers. The results showed that this universal gripper is able to provide a 22% increase in gripping efficiency. Moreover, compared to using multiple grippers and tool changers to handle different products, this one-fits-all gripper has the potential to help users save on operating costs by up to 36%. Manufacturing, Assembly, Automation & Robotics, Infocomm, Robotics & Automation

Current state-of-the-art lab-scale methods for fabricating superhydrophobic membranes for membrane distillation often involve complex surface modifications or the use of nanomaterials. However, these methods are difficult to scale up. This technology relates to a pure rheological spray-assisted non-solvent induced phase separation (SANIPS) approach to fabricate superhydrophobic polyvinylidene fluoride (PVDF) membranes. The resulting membranes have high porosity, superhydrophobicity, high liquid entry pressure, and hierarchical micro/nanostructures. They can also be easily scaled up. The spraying step caused local distortion of the membrane surface, which induced a two-stage phase inversion. This led to the formation of multilevel polymeric crystal structures. The morphological structures and other membrane properties (e.g., mechanical strength and liquid entry pressure) could be tuned by applying spraying materials with different physicochemical properties. This facile fabrication method will pave the way for the large-scale production of superhydrophobic membranes for membrane distillation. Flat sheet membrane: Fabricated from commercial PVDF polymer. Superhydrophobic. High liquid entry pressure. One-step fabrication of the membrane with online modification of the membrane surface. Modules: Industrial-scale modules available. Customized modular design. Spiral-wound modules. Treatment of high salinity waters from mining, metal treatment, pharmaceutical, chemical synthesis, and oil and gas operations. Achieve zero-liquid discharge (ZLD) in industrial processes. Desalination of seawater or brackish water. Treat brine that is produced as a byproduct of desalination. Membrane distillation (MD) is a membrane technology that uses the vapor pressure gradient across a porous hydrophobic membrane to separate water from other components. MD offers several advantages over other membrane separation processes, including: Lower operating pressures Insensitivity to feed concentration for seawater desalination Almost 100% rejection of solutes Relatively low operating temperatures These advantages have led to promising results in MD processes for zero-liquid discharge, desalination, desalination brine treatment, and many other wastewater treatment applications. However, the commercialization of MD has been constrained by the lack of commercially available high-performance MD membranes and high energy consumption. This work addresses the lack of commercially available high-performance MD membranes and has the potential to be the next workhorse of the water industry. Treatment of difficult streams which is not possible with other conventional methods Usage of waste heat High surface area to volume ratio compared to the plate and frame membrane distillation as the current work is in the spiral-wound configuration Proven method of translating membrane fabrication from lab-scale to industrial-scale phase inversion (PI) casting line Readily available industrial-scale process settings to fabricate membrane of one meter in width and several hundreds of meter in length. Membrane Distillation, wastewater treatment Environment, Clean Air & Water, Filter Membrane & Absorption Material

Osteoporosis is a significant global public health concern affecting approximately 500 million people. The condition is associated with high mortality and disability rates due to osteoporotic fractures. The management of osteoporotic fractures comes at a considerable cost of SGD 11K per patient in Singapore, placing a growing burden on healthcare budgets as the aging population increases. Currently, osteoporosis is assessed by measuring bone mineral density (BMD) using dual energy X-ray absorptiometry (DXA). However, the availability of DXA machines, particularly in developing countries, is limited. Consequently, DXA examinations are not routinely ordered, resulting in orthopaedists often lacking DXA results during examinations. Therefore, an alternative method for estimating and screening osteoporosis is necessary. To address this, an automated AI system that can predict a patient's osteoporotic score by evaluating the CTI (cortical thickness index) from a plain femur X-ray scan is designed and developed. This system would provide a preliminary assessment and enable mass screening for osteoporosis. The technology consists of an AI-based software algorithm, to analyse a plain femur X-Ray scan that produces an osteoporosis score, which is equivalent to BMD score produced by DXA machine. The ideal collaboration partners include hospitals and clinics for osteoporosis primary screening, and X-Ray equipment manufacturers to license and translate our invention as an additional feature in their offerings. The proposed automated osteoporotic score prediction technology using AI can be deployed in healthcare industry, e.g. hospitals, medical equipment manufacturers. The technology can be packaged as a cloud-based services, for doctors to use the service for osteoporosis screening from anywhere. It can also be a stand-alone software. The proposed technology is more accessible and less expensive over the traditional method, as it is based on a plain X-ray scan only, which is routinely ordered. The proposed technology is fully automated as it is trained using large amount of data, thus the result is more objective and consistent. The proposed technology can produce osteoporosis score, which is equivalent to BMD score, well-understood by doctors. Infocomm, Artificial Intelligence

This technology offers an automated diagnostic solution for retinal health based on fundus image and deep learning technology. The network automatically classifies fundus images of age-related macular degeneration (AMD), diabetic retinopathy (DR), glaucoma and normal into abnormal and normal classes. The network also can be run on any computing platform, delivering instant results for clinicians and patients. The developed 10-layered neural network can automatically classify images of age-related macular degeneration (AMD), diabetic retinopathy (DR), and glaucoma as abnormal and illustrations of normal subjects as normal. The input image for the system is of size 180 x 270 pixels. The network uses different-sized kernels to interpret the input fundus image, after that, the feature maps are concatenated for analysis. The system was developed and tested on a total of 2986 images (collected from various sources). 'ADAM' optimizer was used to train the net and achieved an accuracy of 95.24% on a set of 1492 images. A system and method for automated retinal health screening using the deep learning CNN technique is developed. The system automatically classifies images of age-related macular degeneration (AMD), diabetic retinopathy (DR), and glaucoma as an abnormal class and images of normal subjects as a normal class. The CNN entails three main layers, the convolution, pooling, and fully connected layers, with a series of convolution and max pooling steps to provide an accuracy of 96.31%, sensitivity of 97.96%, and specificity of 92.67%. The developed network is commercially ready to deploy to any computing or mobile device.   This automated diagnosis solution can be deployed at any clinical facility for the mass screening and routine screening of the fundus. The benefits of the technology include: The diagnosis is fast and reliable. Reduce clinician's workload. Network is compact (small). Readily to be deployed on any computing or mobile device. Healthcare, Telehealth, Medical Software & Imaging

Diabetes is associated with macrovascular and microvascular complications, including Diabetic Foot Ulcers (DFU). To identify and manage DFU risk, diabetic patients are recommended to go for a regular foot assessment. Patients who are at‐risk diabetic foot should undergo regular podiatry evaluation, however specialised diabetes centers are currently facing high rates of ulcer recurrence. Frequent visits to these centers can strain an already overwhelmed healthcare system. The technology developer has invented an Artificial Intelligence (AI) model that is able to detect pre-ulceration. By detecting feet at risk of developing DFU, the model is able to refer patients for timely intervention before it becomes a DFU. Users only need to submit photos of their feet from different angles and an anomaly score will be calculated. The Artificial Intelligence (AI) model is trained to detect pre-ulceration Level of risk can be determined and reflected as scores Able to detect the class of anomalies and classification of data can be modified in the future Hospitals / Clinics Medical Device Manufacturers Pharmaceutical Companies Insurance Providers   Serve as first level of screening for the users – allowing more frequent evaluation without overwhelming the healthcare system Enable self monitoring Semi-supervised approach for AI model training On-device inference providing increased privacy and security Easy-to-use cross platform mobile application Healthcare, Telehealth, Medical Software & Imaging

The technology developer has designed a mobile-friendly Smart Cloud-based inventory solution for users who prefer to access real-time inventory status, such as inventory transactions and inventory levels, and perform simple transactions, on the go.  Equipped with robust analytical capabilities, the solution is capable of providing data-driven recommendations based on the inventory data such as sales trends and order history. The solution is based on open-source platforms such as Google Sheets, AppSheets and Looker Studio. The solution is quick to set up and easy to implement with customisable dashboards and data columns to suit different needs. Staff can also be trained to perform simple customisation of the inventory solution for a company’s unique application. Together with an integrated demand forecasting and re-ordering support system, this solution can help businesses to effectively manage their inventory levels and optimize their supply chain. This technology offer is ideal for businesses seeking a cost-effective cloud-based inventory solution with analytical capabilities to facilitate data-driven decisions. Real-time inventory transactions and monitoring from desktop and mobile devices Open-source solution utilising Google Cloud, Google Sheets, AppSheet and Looker Studio Analytics dashboard provides insights into inventory movement, demand trends and forecasting, as well as comparisons between brands/SKUs/customers/etc. Robust decision-making support for inventory reorder Highly customisable features, including data columns, functions and tasks automation such as WhatsApp approval request This inventory solution is used to help businesses achieve real-time inventory visibility and optimise inventory holdings. This includes start-ups, wholesalers, logistics service providers, manufacturers and e-commerce sellers. The solution benefits companies who wish to own an easy-to-use inventory management system capable of performing inventory transactions, demand forecasting and reorder recommendations. Companies interested in implementing or/and customising the system for internal use may also send their staff for a training course conducted by the technology provider. At the end of the course, the company may obtain a license to use the system for internal use. Quick setup without costly customisation and long lead time Customisable performance dashboards, data tables and inventory processes Easy to trial and cost-effective System control lies wholly with company as solution is customisable and can be maintained in-house by trained staff Infocomm, Cloud Computing

Offshore land reclamation has been an important strategy for Singapore to meet its land needs. However, the ultra-soft soil in the surrounding waters makes land reclamation extremely difficult. Besides, many infrastructure projects (i.e., tunnelling, deep excavation, etc.) are also challenging when encountering soft marine clay due to its poor engineering properties, such as high water content, high compressibility, and low shear strength. Currently, ordinary Portland cement (OPC) is the most common binder used for soft clay stabilisation through deep mixing or jet grouting. However, OPC is not very effective for the stabilisation of marine soft clay with high water content. In addition, the production of OPC leads to negative environmental impacts such as non-renewable resources, high energy consumption, and high carbon emissions. The technology owner has developed a sustainable novel binder, entirely from industrial by-products, that has high stabilisation efficiency for marine soft clay. Using the same binder content, the 28-day strength of the novel binder-stabilised soft clay can be 2–3 times higher than that of the OPC-stabilised clay. In addition, the novel binder has a lower cost and less environmental impact, making it an economical and sustainable alternative to OPC. This technology is available for R&D collaboration, IP licensing, and test-bedding with industrial partners in the construction and infrastructure sectors. The features of this technology are: Renewable sources: entirely from industry by-products High strength: the 28-day strength is 2–3 times higher than that of OPC-stabilised soft clay Low permeability: one order of magnitude lower than that of OPC-stabilised soft clay Cost-effective: the total binder cost is 30–40% lower than that of OPC Low energy consumption: about 70% lower than that of OPC production Low carbon emissions: about 90% lower than those of OPC production The novel binder can be used in deep mixing and jet grouting processes for a variety of construction and infrastructure projects to improve the strength and stability of soft clay. The potential applications are as follows: Densification of granular soils Underground tunnelling Support for deep excavations Underpinning of existing foundations Settlement control Liquefaction mitigation The technology offers the following unique features: Extremely high stabilisation efficiency Low binder cost (30–40% lower than OPC) Renewable resources (from industrial by-products) Low energy consumption and CO2 emissions Easy adaptation to existing soil stabilisation processes This technology is available for R&D collaboration, IP licensing, and test-bedding with industrial partners in the construction and infrastructure sectors. Sustainable Binder, Sofy Clay Stabilization, Deep Mixing, Jet Grouting Materials, Composites, Sustainability, Circular Economy

Singapore commits to achieving net zero emissions by 2050 as part of the nation’s Long-Term Low-Emissions Development Strategy. This target is dependent on many factors including industry’s ability to collect energy consumption data effectively on activities that generate greenhouse gas emissions. Developed by a local research team, this solution provides a cost-effective way to collect consumption data on two key contributors to carbon footprint, electricity and water. Long-range (500m) plug-and-play wireless sensors integrated with a customized dashboard provide users with an intelligent and accurate overview of their utility consumption in their facility. Coupled with an analytics engine running in the background, usage patterns can be established, optimized and contingency alerts provided where required. The solution caters well to facilities with legacy systems by allowing consumption data to be collected without the need for extensive system changes. The system is fully customizable and configurable based on user requirements. The technology owner is looking for partners and collaborators to further co-develop and trial this technology.  Long-range (>500m) wireless connection enables wider, easier installation. Ability to integrate with some models of existing conventional meters Values received are shown in CO2-Eq (1 m3 of “tap water at consumer” to 1.30 kg CO2-Eq). Real-time monitoring of water consumption, leakages, and others. Continuous data collection to allow the study of usage patterns. Data visualization and in-depth analytics. Ability to pinpoint and isolate water-leaking parts SMS Notification system to reduce time wasted before action is taken Ability to shut off remotely using a solenoid valve to reduce water wastage.  Helps to raise general awareness of CO2-Eq Allows savings on CO2-Eq which will quadruple (based on carbon credit) in the next five years for factories Sustainability reporting for listed or IPO companies. (SGX recommendation for core metrics includes GHG emissions, Energy consumption and Water consumption) Factories and Manufacturing Facilities: Ideal for wastewater management, water-flow and additional IoT sensors may be added to track that the output remains within regulations, or recycled inputs are within specifications. Idle machines can also be switched off, allowing better resource management. Public Institutions: Complements and reinforces water conservation efforts Private Commercial Facilities- provides surveillance for a long-term redesign of water distribution throughout a facility for cost-effectiveness. Housing facilities: Monitor the usage of electricity and water to study possible areas of savings to reduce CO2-Eq. Help to raise general awareness regarding carbon footprint generation. Facilities with unstable power supply or critical processes: With the monitoring and control system, it is possible to control the usage of resources when it is limited or restricted. Power to non-critical equipment may be cut to reduce consumption, and the allocation of fluids may also be set based on different conditions. Compared to existing systems in the market which typically cater to either power or energy monitoring, this solution offers the additional integration of water monitoring as well as additional control features to trigger intervention steps should there be anomalies such as piping leaks. The system allows direct insight to the correlation between what is used to carbon footprint. Data is collected to help users understand the greatest generators of carbon footprint, so that users are more aware, and can consider ways to reduce their carbon equivalent emissions. Electronics, Sensors & Instrumentation, Green Building, Sensor, Network, Building Control & Optimisation, Sustainability, Low Carbon Economy