AI Research

Artificial Intelligence in Cataract Surgery and Optometry at Large with Harvey Richman, OD, and Rebecca Wartman, OD

Published

2 months ago

July 9, 2025

At the 2025 American Optometric Association Conference in Minneapolis, MN, Harvey Richman, OD, Shore Family Eyecare, and Rebecca Wartman, OD, optometrist chair of AOA Coding and Reimbursement Committee, presented their lecture on the implementation of artificial intelligence (AI) devices in cataract surgery and optometry at large.¹

AI has been implemented in a variety of ophthalmology fields already, from analyzing and interpreting ocular imaging to determining the presence of diseases or disorders of the retina or macula. Recent studies have tested AI algorithms in analyzing fundus fluorescein angiography, finding the programs extremely effective at enhancing clinical efficiency.²

However, there are concerns as to the efficacy and reliability of AI programs, given their propensity for hallucination and misinterpretation. To that end, Drs. Richman and Wartman presented a study highlighting the present and future possibilities of AI in cataract surgery, extrapolating its usability to optometry as a whole.

Richman spoke to the importance of research in navigating the learning curve of AI technology. With the rapid advancements and breakneck pace of implementation, Richman points out the relative ease with which an individual can fall behind on the latest developments and technologies available to them.

“The problem is that the technology is advancing much quicker than the people are able to adapt to it,” Richman told HCPLive. “There’s been research done on AI for years and years; unfortunately, the implementation just hasn’t been as effective.”

Wartman warned against the potential for AI to take too much control in a clinical setting. She cautioned that clinicians should be wary of letting algorithms make all of the treatment decisions, as well as having a method of undoing those decisions.

“I think they need to be very well aware of what algorithms the AI is using to get to its interpretations and be a little cautious when the AI does all of the decision making,” Wartman said. “Make sure you know how to override that decision making.”

Richman went on to discuss the 3 major levels of AI: assistive technology, augmented technology, and autonomous intelligence.

“Some of those are just bringing out data, some of them bring data and make recommendations for treatment protocol, and the third one can actually make the diagnosis and treatment protocol and implement it without a physician even involved,” Richman said. “In fact, the first artificial intelligence code that was approved by CPT had to do with diabetic retina screening, and it is autonomous. There is no physician work involved in that.”

Wartman also informed HCPLive that a significant amount of surgical technology is already using artificial intelligence, mainly in the form of pattern recognition software and predictive devices.

“A lot of our equipment is already using some form of artificial intelligence, or at least algorithms to give you patterns and tell you whether it’s inside or outside the norm,” Wartman said.

References

Richman H, Wartman R. A.I. in Cataract Surgery. Presented at the 2025 American Optometric Association in Minneapolis, MN, June 25-28, 2025.
Shao A, Liu X, Shen W, et al. Generative artificial intelligence for fundus fluorescein angiography interpretation and human expert evaluation. NPJ Digit Med. 2025;8(1):396. Published 2025 Jul 2. doi:10.1038/s41746-025-01759-z

Source link

Up Next

Wiley Partners with Anthropic on AI Integration for Scholarly Research

Don't Miss

Artificial intelligence could hire you. Now it could also fire you

Harvey Richman, OD

Click to comment

AI Research

Chinese social media firms comply with strict AI labelling law, making it clear to users and bots what’s real and what’s not

Published

21 minutes ago

September 2, 2025

Jon Martindale

Chinese social media companies have begun requiring users to classify AI generated content that is uploaded to their services in order to comply with new government legislation. By law, the sites and services now need to apply a watermark or explicit indicator of AI content for users, as well as include metadata for web crawling algorithms to make it clear what was generated by a human and what was not, according to SCMP.

Countries and companies the world over have been grappling with how to deal with AI generated content since the explosive growth of popular AI tools like ChatGPT, Midjourney, and Dall-E. After drafting the new law in March, China has now implemented it, taking the lead in increasing oversight and curtailing rampant use with its new labeling law making social media companies more responsible for the content on their platforms.

Chinese officials claim the law is designed to help combat AI misinformation and fraud, and is applicable to all the major social media firms. That includes Tencent Holdings’ WeChat – a Chinese WhatsApp equivalent – which has over 1.4 billion users, and Bytedance’s TikTok alternative, Douyin, which has around a billion users of its own. Social media platform Weibo, with its 500 million plus active monthly users, is also impacted, as is social media and ecommerce platform, Rednote.

Although China is the first major country to implement an AI content labeling system through legislation, similar conventions are being considered elsewhere, too. Just last week the, Internet Engineering Task Force proposed a new AI header field which would use metadata to disclose if content was AI generated or not. That wouldn’t necessarily make it easier for humans to tell the difference, but it would give the algorithms a heads up that what it’s viewing may not be human-crafted.

Source link

AI Research

RSC partners with Enago for AI-powered manuscript screening

Published

50 minutes ago

September 2, 2025

The Editors

Shutterstock.com/LALAKA

The Royal Society of Chemistry (RSC) has partnered with publishing solutions company Enago to deploy bespoke artificial intelligence (AI) technology for screening incoming journal submissions. The move is designed to streamline the manuscript submission process and enhance the author experience.

The collaboration will see Enago’s AI-powered manuscript screening technology, built on the company’s Enago Reports platform, integrated into RSC’s submission workflow. The system will check manuscripts against journal-specific requirements across RSC’s portfolio, providing authors with targeted guidance on compliance before submission.

“We are excited to partner with Enago to deliver this innovative pre-submission tool to support our authors with manuscript preparation and to speed up the submission process,” said Emma Wilson, Director of Publishing at RSC. “Investing in author tools such as this supports our goal to provide an enhanced and excellent author experience.”

The technology represents a shift towards automated pre-submission checking, addressing a common pain point in academic publishing where manuscripts are frequently rejected or delayed due to formatting and compliance issues rather than scientific content.

Abhigyan Arun, CEO of Enago said: “The RSC is one of the foremost publishers progressing high quality scientific research and it is a privilege to partner with them,” he said. “Providing authors with accurate actionable information on their manuscript is a step towards improving editorial and peer review efficacy.”

The system is designed to reduce the administrative burden on editorial teams while helping authors prepare submissions that meet specific journal requirements from the outset. This could potentially reduce submission-to-decision times and improve overall publishing efficiency.

The partnership also reflects a broader trend in academic publishing towards AI-assisted manuscript processing, as publishers seek to balance efficiency gains with maintaining rigorous peer review standards.

Earlier this year, Enago announced the launch of DocuMark, developed by Trinka AI. DocuMark is a platform designed to transform how academic institutions address AI-assisted student submissions, shifting the focus from detection to transparency.

Source link

AI Research

Humanoid robots lack data to keep pace with explosive rise of AI

Published

54 minutes ago

September 2, 2025

Atharva Gosavi

Greece recently witnessed the world’s first International Humanoid Olympiad in Olympia, where humanoid robots played boxing and soccer matches to attain glory.

The event, held from August 29 to September 2, was organized by Acumino and Endeavor, who invited industry leaders to line up as speakers, apart from the smart machines displaying their abilities.

While humanoid robots have increasingly gained popularity for mirroring human actions, we have yet to see them involved in routine household chores like washing dishes and tidying closets.

Comparisons with AI

AI has advanced explosively in the past year through applications like ChatGPT, but the same cannot be said about its physical cousins – the humanoid robots. Humanoid robots are miles behind in learning from data compared to AI software and tools.

Minas Liarokapis, a Greek academic and startup founder who organized the Olympiad, made a rather bold prediction regarding humanoids becoming a helping hand in the kitchens and other household chores.

“I really believe that humanoids will first go to space and then to houses … the house is the final frontier,” she told the Associated Press (AP) on Tuesday.

“To enter the house, it’ll take more than 10 years. Definitely more,” said Liarokapis.

“I’m talking about executing tasks with dexterity, not about selling robots that are cute and are companions,” she continued.

Pinpointing the AI advantage

Any AI tool or software needs vast data for training to perform at its best. Fortunately, there’s colossal data available for training with such tools. The same, however, cannot be said for humanoids and robots.

Humanlike robots are roughly 100,000 years behind AI in learning from data, all thanks to that large divide in data availability.

Ken Goldberg, a University of California, Berkeley professor, devised a novel solution to bridge this gap. He has urged makers to go beyond simulations and make robots “collect data as they perform useful work, such as driving taxis or sorting packages.”

As it happens, researchers and scientists are already using reinforcement learning as a means to help humanoid robots learn from data in real time. This technology has helped them save valuable time by programming the machines for every action at every step.

Developing a robotic brain

The Olympiad event also hosted Hon Weng Chong, CEO of Cortical Labs, as one of the esteemed personalities in the lineup of speakers.

Chong revealed that his biotech company is developing a biological computer brain that will learn like humans.

This brain uses real brain cells grown on a chip for learning from data. These cells can learn and respond to information at a faster rate, helping robots think and adapt like humans.

The dire need for faster robotic learning

At the Humanoid Olympics, organizers focused on realistic challenges to ensure fair progress checks. Co-founder Patrick Jarvis noted that while events like discus or javelin were considered, they proved too complex.

High jump was also ruled out due to the need for specialized legs. Instead, competitions highlighted tasks that humanoid robots could practically achieve, ensuring meaningful demonstrations of capability.

However, those limitations are also a stark reminder of why faster learning is essential for humanoid robots to rival the rise of AI software and tools. Bridging that gap will decide whether humanoid robots remain niche performers or evolve into everyday companions alongside advanced AI.

Source link