Latest news with #ConvolutionalNeuralNetwork
The Star
26-05-2025
- Science
- The Star
Leaf it to tech: Are AI-powered apps giving plants a voice?
The idea of conversing with greenery has long captured the human imagination, from the giant talking trees popularised in J.R.R. Tolkien's The Lord Of The Rings to the whimsically poetic The Lorax from Dr Seuss. And with advances in technology, the possibility of that happening in real life might not be so far off. The artificial intelligence (AI) boom that kicked off near the end of 2022 gave rise to countless mobile apps purporting to act as gardening assistants capable of identifying the exact species of plants from just a photo and 'speaking' on their behalf. While not quite at the level of 'I speak for the trees' à la the fictional Lorax character, such apps can be quite handy in a pinch for novices who can't tell grass from unwanted weeds, with many having been installed millions of times with largely positive ratings on app stores. According to Prof Vinesh Thiruchelvam, chief innovation and enterprise officer at Asia Pacific University (APU), these apps typically use AI-based Computer Vision and Image Processing to identify plant types. Prof Vinesh points out that these apps struggle in recognising less common species, citing only a 60% accuracy when it comes to unique lettuce varieties, while at the same time also losing out on local context. — VINESH THIRUCHELVAM 'The AI imagery database is trained, and a Convolutional Neural Network is utilised for analysing images. It works by breaking the image into small pieces and looking for patterns. 'These patterns help it recognise things in the image, like edges or shapes of different plants,' he says, adding that apps then use this information to provide real-time basic care advice. Growing a green thumb From Dr Nallammai Singaram's perspective, such apps make it much easier for those just starting with their home garden. A senior lecturer at the School of Biosciences, Faculty of Health & Medical Sciences at Taylor's University, she finds that 'these apps have made plant care less intimidating'. 'If I were putting a plant in my house for the first time, I wouldn't know what sort of soil to use, how to pot my plant, or how to make a potting mix. 'I might not know these things, but these apps would give you that kind of guide,' Nallammai says, adding that in the case of a rose plant, for example, an app may suggest the ideal soil mix ratio and recommend planting it about six inches below the surface. Prof Alan Dorin from the Faculty of Information Technology at Monash University shares similar thoughts, noting these apps put a wealth of information at the fingertips of those first getting started, making gardening more accessible and removing the need for extensive prior knowledge. 'For interested and engaged people, this knowledge was previously accessible via traditional means – talking to friends and neighbours, the staff at the local plant nursery, or reading gardening books, magazines, watching TV or listening to gardening radio programmes. 'The immediacy of access to information related to a specific plant is what makes these apps a stand-out method for increasing engagement with gardens,' he says. Prof Dorin notes that these apps put a wealth of information at the fingertips of those first getting started, making gardening more accessible and removing the need for extensive prior knowledge. — ALAN DORIN Nallammai says this kind of guidance can boost beginners' confidence, help avoid common mistakes, and turn gardening into a more enjoyable and therapeutic experience. 'If you truly enjoy gardening, you naturally become curious and eager to learn more about your 'green babies', exploring every way to help your plants thrive. 'This curiosity, supported by apps and digital tools, can positively influence gardening habits,' she says. Back to earth However, Nallammai also stresses that while they are a good starting point, it is 'important to remember that apps alone cannot make you an expert; real learning also requires hands-on experience and personal observation'. 'For some, the apps spark long-term interest and lasting gardening habits and bring joy. For others, they are novelty tools that provide convenience at the beginning but don't replace the need to develop real gardening knowledge,' she says. Prof Vinesh agrees, saying that 'apps are a wonderful digital support tool. But at times, it does not consider local conditions like real-time weather, soil, or heat'. 'That's why they can't replace real hands-on gardening experience, especially for serious gardeners or farmers who have to deal with the volatile climate change issues or the growing new patterns of pest attacks.' Nallammai believes the takeaway shouldn't be to entirely avoid using these apps, but rather to not rely solely upon them. — SAMUEL ONG/The Star Apps still do have a place, from Nallammai's experience, having observed her students using these kinds of apps to quickly identify plant species and receive recommendations for improving the growth of the test plants they are working with in class. She says apps like PictureThis , PlantNet , and Seek By iNaturalist are reliable for plant identification, while Planta is helpful for watering schedules, and care reminders. Plantora , on the other hand, not only detects nutrient deficiencies and signs of plant disease but also provides possible solutions to address them. It is worth noting though that most of these apps require a paid subscription to unlock their full functionality. Meanwhile, Prof Dorin says he 'was recently working with a group of primary school children who had started using apps to identify plants in their school yard'. 'Their goal was to learn whether the plants provided good resources for native bees. Their excitement and interest were infectious, heart-warming to see. 'Likewise, I know some retirees who are using these AI tools on their daily walks in local parks and gardens to identify plants and flowers they encounter. The apps provide a virtual 'guide' that is easy to appreciate,' he says. Barking up the wrong tree However, like other AI tools, these apps aren't always entirely accurate, so it's important to be cautious when receiving suggestions. Prof Vinesh points out that these apps struggle in recognising less common species, citing only a 60% accuracy when it comes to unique lettuce varieties, while at the same time also losing out on local context. From Nallammai's testing, this appears to be true. She highlights that the datasets used to train these AI apps may come from different parts of the world, which means they may not be entirely relevant or accurate for local plant species and conditions. 'I tested a yellow-green philodendron with multiple apps: PictureThis identified it as a lemon lime philodendron, another app called it a giant philodendron, while in PlantIn , they already had an existing lemon lime philodendron with different features. 'Different apps have different capabilities. Some of the rarer species may not be included in their databases, so the app might not be able to detect them. That's one of the limitations,' she says. These differences may seem minor, but Nallammai says they can indicate what specific kind of care a plant needs. The artificial intelligence (AI) boom that kicked off near the end of 2022 gave rise to countless mobile apps purporting to act as gardening assistants capable of identifying the exact species of plants from just a photo and 'speaking' on their behalf. — Image by freepik 'The darker green ones can capture better sun and may need more sunlight, the lighter ones would prefer to be a little bit on the shadier side. 'The amount of water you put might differ. The thickness of the plant, leaf thickness will also change; therefore, if it is exposed to too much sun, it will get burnt, so maybe less sun. 'They're from the same species but different variations, which could be a random mutation, or from people intentionally breeding them,' she says. Nallammai believes the takeaway shouldn't be to entirely avoid using these apps, but rather to not rely solely upon them. She encourages budding gardeners to start reading up and building their knowledge while engaging with the gardening community, especially more experienced gardeners who can offer advice from their experience. Prof Dorin similarly says: 'If there's a shortcoming, it probably lies in people's over-reliance and over-confidence in the technology. 'The ability of an AI tool to make an informed recommendation will always be limited by the amount and quality of data it receives in training, and the data it is given when asked to analyse a specific situation. 'For example, it would be tricky for AI software to give reliable and specific care information regarding a plant it hasn't been exposed to before. 'But the AI may nevertheless mislead a human by confidently offering advice based on poor evidence,' he adds. Speaking for the trees While no one should expect their sunflowers to suddenly start calling them by name or demanding more water and sunlight, the way Nallammai sees it, plants are 'talking' in their own way, though, in a language that humans don't quite understand. This is the gap that projects like the Avanade Intelligent Garden in the United Kingdom aim to bridge, allowing people to learn about the individual needs of each tree through one-on-one interactions powered by smart sensors and AI technology. From smart sensors to AI apps, technology is giving green thumbs deeper insight into what their plants need to thrive. — Image by freepik According to Nallammai, similar sensors have been deployed at Taylor's Urban Farm to monitor soil electrical conductivity and pH levels (acidity and alkalinity), helping ensure that the hydroponic system delivers the right amount of water and nutrients for healthy plant growth. A hydroponic system is a method of growing plants without soil that delivers water, oxygen, and nutrients directly to the plants. She says such systems are available at significantly smaller scales for home gardening use for as low as RM50 for a system with basic monitoring capabilities via an app, and can rise to the RM500 range for more sophisticated options with more sensors and more detailed monitoring. For casual gardeners just starting out, simple apps paired with an online kit can provide basic monitoring and easy-to-follow guidance. An app that sends reminders for watering or fertilising would also be helpful for beginners in caring for their plants without overwhelming them. Meanwhile, those looking for a challenge and wanting to explore might want more comprehensive sensor and irrigation systems to cultivate rarer species or achieve greater yields and productivity. 'The moment the plant is stressed, it's telling you what it needs in its own language,' Nammallai says. 'A very simple example: when you go home today and see that your plant is wilted. You know your plant is crying out for water. 'So if I have an app that can sense the turgidity of the leaves, it can capture that information and tell you the leaves aren't turgid, meaning they're lacking moisture. That's the plant indirectly telling you it needs water,' she says. Nammallai adds that this is already happening today with the wide array of sensors that can detect compounds in the plants, growth regulators, proteins, light, and other elements. If the right sensors are in place, each one can provide a piece of the overall picture. Nallammai explains that by combining sensor data with visual observations and other morphological indicators, it becomes possible to link them together and begin forming a kind of language that translates plant signals into meaningful insights for growers. 'Using these chemicals in the plant, we can decipher what it is 'saying'. 'The plant might just turn around and tell you, 'you know what, I'm ready, I'm heading towards maturity, I'm going to bear fruits, just hang in there',' she says.
Time of India
30-04-2025
- Health
- Time of India
Class X boy's AI model to detect spinal lesion; wins Rs 1.5L
Lucknow: Class X student Visharad Upadhyay , who created an AI model to detect and classify spinal lesion using Convolutional Neural Network (CNN), won Rs 1.5 lakh in Hackofiesta 6.0 organised by UP govt and IIIT Lucknow, on Wednesday. Tired of too many ads? go ad free now Student of Kendra Vidyalaya Gomtinagar, Visharad said, "Spinal lesions are difficult to detect in X-rays with human eyes and often require an MRI or CT scan. Since there are not many diagnostic centres equipped with MRI and CT scan in rural and semi-urban areas, my AI model prototype can detect lesions in the spine from X-ray images and it's highly accurate." Indranagar resident and son of retired sergeant, the tech-savvy teenager said, "The model has been designed to target seven types of lesions including osteophytes, disc space narrowing, surgical implant, foraminal stenosis, spondylolisthesis, vertebral collapse." Visharad started coding at the age of 10, while his elder sister Priya Upadhyay was pursuing her master's in computer application. "I was inspired by my elder sister. I spend considerable time coding and often get checked by my family to focus on the Class X board, but I'm able to manage it." "After school, I want to join a university which can help me learn more about computer science and AI," he added. Visharad's father Dinesh Kumar Upadhyay now practices law after retiring from IAF, while mother Alka Upadhyay is a retired teacher.
Yahoo
23-02-2025
- Health
- Yahoo
AI-driven blood analysis may spur cancer detection, Hong Kong prenatal test pioneer says
Affordable cancer screening may be available to Hong Kong residents within three years, as advances in artificial intelligence (AI) make the early detection of malignant tumours easier and faster, according to the creator of a pioneering prenatal test for Down syndrome. AI helps researchers recognise patterns in the gene fragments they find in blood plasma, sparking a revolution in the field of epigenetics, or the study of non-mutation behavioural effects on genes, said Professor Dennis Lo Yuk-ming. The AI-powered method enables researchers to decipher epigenetic signals from DNA samples without the damaging effects of the chemical treatment that destroys 90 per cent of the genetic material, he said. Do you have questions about the biggest topics and trends from around the world? Get the answers with SCMP Knowledge, our new platform of curated content with explainers, FAQs, analyses and infographics brought to you by our award-winning team. "My team members and I were asking: is it possible to use some of the AI methods developed for facial recognition [to help us] see things we cannot see?" Lo said during a January 7 interview with the Post. "As it turned out, it is." Professor Allen Chan Kwan-chee (left) and Professor Dennis Lo Yuk-ming (right) of the Chinese University of Hong Kong on 10 August 2017. Photo: Dickson Lee alt=Professor Allen Chan Kwan-chee (left) and Professor Dennis Lo Yuk-ming (right) of the Chinese University of Hong Kong on 10 August 2017. Photo: Dickson Lee> Lo's team at the Centre for Novostics at the Hong Kong Science and Technology Park developed an in-house AI system based on the Convolutional Neural Network, a deep learning model that was designed mainly for image recognition. With advances in generative AI tools such as ChatGPT, the company also incorporated the transformer neural network architecture into the system, he said. Like many biomedical companies at the forefront of drug discovery, diagnostics and treatment, Novostics is diving into the use of AI. Like many companies in Hong Kong and mainland China, the company "has faced challenges" in accessing computing power because of US curbs on access to cutting-edge AI chips. Professor Dennis Lo Yuk-ming with a chip that could sequence up to 8 million single DNA molecules for genomic research and genetic disease detection, at the Prince of Wales Hospital in Shatin on 20 October 2020. Photo: K. Y. Cheng alt=Professor Dennis Lo Yuk-ming with a chip that could sequence up to 8 million single DNA molecules for genomic research and genetic disease detection, at the Prince of Wales Hospital in Shatin on 20 October 2020. Photo: K. Y. Cheng> "Fortunately, we still have a pretty good system that we developed over the years", said Lo, a day before he formally took over as president of the Chinese University of Hong Kong (CUHK). "With the difficulty in getting the latest generation of Nvidia chips, we just have to try other chipmakers that can still provide what we need." Lo's team is developing tests for the early detection of cancers, using a similar method as his pioneering work on using the blood plasma of pregnant women to check for Down syndrome and other genetic disorders. Nasopharyngeal cancer (NPC), a throat tumour that is common in southern China's Guangdong province, was the first on Lo's target list. Using the cancer's suspected link with the Epstein-Barr virus, Lo developed a test in 2013 for finding it in blood plasma. Single-use DNA cells for genome sequencing, at the Prince of Wales Hospital in Shatin on 20 October 2020 . Photo: K. Y. Cheng alt=Single-use DNA cells for genome sequencing, at the Prince of Wales Hospital in Shatin on 20 October 2020 . Photo: K. Y. Cheng> After collecting 20,000 samples over four years, the study concluded in July 2023 that screening for the virus was an effective detection tool for the cancer, helping to slash the mortality rate to a mere 3.7 per cent from 40 per cent, according to Lo. The test is now available in Hong Kong for HK$1,500 (US$192.5). "Even though Hong Kong's healthcare system is pretty good, 75 per cent of nasopharyngeal cancer was detected late" before the new test was introduced, Lo said. The use of AI widened the possibilities of finding other cancers even if a virus may not be behind every cancer, because the discovery of each mutation was "like hunting one thing out of a million", he said. Cirina, which Lo founded in Hong Kong after creating the NPC test, merged in 2017 with Grail, a company in California that speciliased in cancer detection. The merged company developed the Galleri test to screen for more than 50 types of virus-related cancers, a product that was named as one of The Best Inventions of 2022 by Time magazine. Grail was bought for US$8 billion in 2021 by the biotechnology giant Illumina, based in San Diego. Its multi-cancer test is only available in the US market for US$949 (HK$7,380), and is not covered by most healthcare coverage plans. Lo's team is trying to make a cheaper version of the test for mainland China and the rest of the world. Novostics has developed a test called Fragma, which can find the DNA fragments of lung cancer cells and liver carcinoma in blood plasma and urine. In 2023, Lo established a US$200 million venture called Insighta with Nasdaq-listed Prenetics Group to commercialise the Fragma tests. Tencent Holdings invested US$30 million last October in Insighta, valuing the start-up at US$200 million. Insighta, based at the Hong Kong Science and Technology Park, is conducting large-scale clinical tests of the Fragma process in Hong Kong and mainland China. It aims to eventually offer the test at US$200 (HK$1,565). Some of the earlier products, such as the test for liver cancer, will "hopefully" be available "in two or three years from now", Lo said. "If we want to do it in a more resource-challenged environment, we need the technology to be cheaper." This article originally appeared in the South China Morning Post (SCMP), the most authoritative voice reporting on China and Asia for more than a century. For more SCMP stories, please explore the SCMP app or visit the SCMP's Facebook and Twitter pages. Copyright © 2025 South China Morning Post Publishers Ltd. All rights reserved. Copyright (c) 2025. South China Morning Post Publishers Ltd. All rights reserved.
