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.
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Forbes
34 minutes ago
- Forbes
What Could Future Banking Look Like If AI Takes Over?
Alex Kreger, UX Strategist & Founder of the financial UX design agency UXDA, designs leading banking and fintech products in 39 countries. getty The imminent integration of AI into daily routines promises to dramatically reshape our lives over the next five years, propelled by advancements akin to ChatGPT, Gemini, Grok, etc. This shift is driven by the recognition that human capacities, while remarkable, cannot match the vast research and creative and analytical potential of artificial intelligence (AI). As we project into the future, it becomes clear that AI will also redefine digital banking experiences and grant individuals with financial capabilities that were once unimaginable. As a design strategist developing financial services for leading banks and fintech providers in 39 countries, I'm curious to envision how AI will overhaul the typical banking experience for everyday consumers. Although the complete adoption of AI across the financial sector has yet to unfold, it is crucial to anticipate its eventual impact. The question is no longer 'What if?' but rather 'How?'—and how best to brace ourselves for the changes that lie ahead. Banks already maintain enormous stores of customer data, but unlocking its true power demands cutting-edge technology. AI may well be the solution that helps institutions tackle customer demands with speed and accuracy. By channeling this data effectively, banks can provide individualized products at precisely the right time—an endeavor impossible for standard processes alone. The current data stockpile is merely a starting point. As digital tools evolve, financial institutions will gather much more data from smartphones, social networks, public service APIs, open banking APIs and IoT devices through 5G. This explosion of information calls for a robust, near-superhuman capacity to sift through the noise and pinpoint what truly matters—something AI might deliver within the next decade. In the coming years, the most significant AI-driven breakthroughs are likely to include: • Personalized Offers: Data-rich approach makes customized proposals more precise and simultaneously mitigates risks by matching the ideal product to the ideal customer. • Investment: By eliminating human biases, AI could evaluate a multitude of market and business variables to foresee investment success. • Security: AI could expedite verification by reducing the constant need for identity confirmations. • Financial Advisory: With the aid of big data and personal profiling, AI could illuminate each client's needs, generating in-depth forecasts and healthier financial practices. • Support: AI-powered bots could offer prompt, tailored solutions, greatly enhancing customer service. • Alternative Processing: AI-powered voice, gestures, neurotechnology, VR and AR interfaces will enable banking transactions beyond conventional channels. With Statista expecting generative-AI spend in banking to rocket to $85 billion by 2030, it's time for leaders to start by putting AI into their strategic plan—not just the tech roadmap. Hire a senior executive (Chief AI) who owns value creation and AI risks and spin up a cross-functional 'AI initiatives' that groups stakeholders, data scientists and product designers that move to an API-first, event-streaming service architecture so models can surface predictions (e.g., 'potential cash shortfall Friday') in real time. Early adopters are showing where the value sits, and leaders should take note. J.P. Morgan's Quest IndexGPT can generate investable indices; Morgan Stanley's Debrief can summarize adviser meetings; NatWest's Cora+ can handle nuanced customer queries. At the same time, Wall Street majors—from Goldman to Citi—are scaling internal LLM-powered co-pilots for drafting IPO documents, surfacing research or searching policies. Customer-facing assistants are already setting the bar. Bank of America's Erica has served 20 million active users, Wells Fargo's Fargo went from 21 million interactions in 2023 to 245 million in 2024 by using a privacy-first pipeline that strips PII before any LLM call. On the insight side, RBC's NOMI Forecast crunches account data to predict the next seven days' cash flow; more than 900,000 clients have generated 10 million interactions since its late-2021 launch. Generative models excel at turning trillions of events into the next best micro-experience. Commonwealth Bank of Australia's Customer Engagement Engine, for example, ingests 3.1 trillion data points and runs 2,000 real-time models, lifting loyalty with recommendations so much that mobile users now log in 67 times a month on average. The key is to couple a real-time feature store with small language models that handle intent, then let a larger model draft the personalized nudge or insight. Start with one or two journeys where better prediction or conversation will be felt within weeks—fraud alerts or an SME cash-flow coach. Ship, measure, retrain and fold the learning into a reusable component library so subsequent squads stand on the shoulders of the first. The biggest headwind is regulation: Europe's AI Act is already in force and will classify credit-scoring, KYC, trading and robo-advice models as 'high risk' by August 2026. Finding talent and culture is also an ongoing challenge. Banks are hiring aggressively, yet even Deutsche Bank admits the scarcity of seasoned AI professionals and the difficulty of embedding them in legacy teams. Third, security and trust: four in five bank leaders say they fear AI-enabled cyberattacks, and front-office chatbots can still hallucinate or breach privacy if left unsupervised. Mitigate by adopting zero-trust data-access patterns, embedding red-teaming into MLOps, and running 'constitutional' or retrieval-augmented QA layers that force a model to cite source documents. Initially, AI's role is to automate foundational tasks. Over time, however, I expect that it will evolve to deliver comprehensive solutions across all industries, including finance. After two decades of digital self-service in finance, AI can restore the conversation—context-aware, always on, and scaled to every customer. AI's full potential is truly immeasurable, and its effects on banking customer experience—and countless other sectors—will be transformative. By merging technological advancements with thoughtful user experience design, forward-looking companies can build a future where AI not only empowers individuals but also redefines entire industries. The era of AI-driven finance is fast approaching, and now is the time to prepare for its far-reaching influence. Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?

Engadget
an hour ago
- Engadget
OpenAI appeals court order forcing it to preserve all ChatGPT data
OpenAI has appealed a court ruling from last month that forces it to retain ChatGPT data indefinitely as part of a copyright violation case brought by The New York Times in 2023. CEO Sam Altman said in a tweet on X that the judge's decision "compromises our users' privacy" and "sets a bad precedent." In May, federal judge Ona T. Wang ordered OpenAI to preserve and segregate all ChatGPT output log data that would otherwise be deleted due to a user request. She said that the ruling was justified because the volume of deleted conversations is "significant." The directive notes that the judge asked OpenAI if there was a way to anonymize the data to address users' privacy concerns. The New York Times sought the order so that it can accurately track how often OpenAI violates its IP, including instances when users requested deletion of chats. A federal judge allowed the original case to proceed, agreeing with the NYT 's argument that OpenAI and Microsoft's tech had induced users to plagiarize its materials. In a FAQ on its site, OpenAI painted the order as a privacy issue without addressing the millions of alleged copyright violations. "This fundamentally conflicts with the privacy commitments we have made to our users," the company wrote. "It abandons long-standing privacy norms and weakens privacy protections." OpenAI noted that the order "does not impact ChatGPT Enterprise or ChatGPT Edu customers." The NYT and other AI copyright cases are still ongoing, as courts have not yet decided whether OpenAI, Google and other companies infringed copyrights on a massive scale by scraping material from the internet. The tech companies have argued that training is protected by "fair use" copyright law and that the lawsuits threaten the AI industry. Creators of that content, in turn, argue that AI harms their own livelihoods by stealing and reproducing works with little to no compensation.


CNET
an hour ago
- CNET
Siri Could Become More Like ChatGPT. But Why?
Siri Could Become More Like ChatGPT. But Why? Siri Could Become More Like ChatGPT. But Why? Click to unmute Video Player is loading. Play Video Pause Skip Backward Skip Forward Next playlist item Unmute Current Time 0:07 / Duration 5:19 Loaded : 14.89% 0:06 Stream Type LIVE Seek to live, currently behind live LIVE Remaining Time - 5:12 Share Fullscreen This is a modal window. Beginning of dialog window. Escape will cancel and close the window. Text Color White Black Red Green Blue Yellow Magenta Cyan Opacity Opaque Semi-Transparent Text Background Color Black White Red Green Blue Yellow Magenta Cyan Opacity Opaque Semi-Transparent Transparent Caption Area Background Color Black White Red Green Blue Yellow Magenta Cyan Opacity Transparent Semi-Transparent Opaque Font Size 50% 75% 100% 125% 150% 175% 200% 300% 400% Text Edge Style None Raised Depressed Uniform Drop shadow Font Family Proportional Sans-Serif Monospace Sans-Serif Proportional Serif Monospace Serif Casual Script Small Caps Reset Done Close Modal Dialog End of dialog window. Close Modal Dialog This is a modal window. This modal can be closed by pressing the Escape key or activating the close button. Close Modal Dialog This is a modal window. This modal can be closed by pressing the Escape key or activating the close button. Siri Could Become More Like ChatGPT. But Why? Siri Could Become More Like ChatGPT. But Why? Jun 6, 2025 Tech Show Transcript Apple faces pressure to deliver improvements to Siri with AI, but leadership changes on the Siri team could lead to different paths for the voice assistant. CNET's Bridget Carey looks at recent rumors and asks how AI can make your life better.