Eyes may be the window to your soul, but the tongue mirrors your health

 A 2000-year-old practice by Chinese herbalists -- examining the human tongue for signs of disease -- is now being embraced by computer scientists using machine learning and artificial intelligence.

Tongue diagnostic systems are fast gaining traction due to an increase in remote health monitoring worldwide, and a study by Iraqi and Australian researchers provides more evidence of the increasing accuracy of this technology to detect disease.

Engineers from Middle Technical University (MTU) in Baghdad and the University of South Australia (UniSA) used a USB web camera and computer to capture tongue images from 50 patients with diabetes, renal failure and anaemia, comparing colours with a data base of 9000 tongue images.

Using image processing techniques, they correctly diagnosed the diseases in 94 per cent of cases, compared to laboratory results. A voicemail specifying the tongue colour and disease was also sent via a text message to the patient or nominated health provider.

MTU and UniSA Adjunct Associate Professor Ali Al-Naji and his colleagues have reviewed the worldwide advances in computer-aided disease diagnosis, based on tongue colour, in a new paper in AIP Conference Proceedings.

"Thousands of years ago, Chinese medicine pioneered the practice of examining the tongue to detect illness," Assoc Prof Al-Naji says.

"Conventional medicine has long endorsed this method, demonstrating that the colour, shape, and thickness of the tongue can reveal signs of diabetes, liver issues, circulatory and digestive problems, as well as blood and heart diseases.

"Taking this a step further, new methods for diagnosing disease from the tongue's appearance are now being done remotely using artificial intelligence and a camera -- even a smartphone.

"Computerised tongue analysis is highly accurate and could help diagnose diseases remotely in a safe, effective, easy, painless, and cost-effective way. This is especially relevant in the wake of a global pandemic like COVID, where access to health centres can be compromised."

Diabetes patients typically have a yellow tongue, cancer patients a purple tongue with a thick greasy coating, and acute stroke patients present with a red tongue that is often crooked.

A 2022 study in Ukraine analysing tongue images of 135 COVID patients via a smartphone showed that 64% of patients with a mild infection had a pale pink tongue, 62% of patients with a moderate infection had a red tongue, and 99% of patients with a severe COVID infection had a dark red tongue.

Previous studies using tongue diagnostic systems have accurately diagnosed appendicitis, diabetes, and thyroid disease.

"It is possible to diagnose with 80% accuracy more than 10 diseases that cause a visible change in tongue colour. In our study we achieved a 94% accuracy with three diseases, so the potential is there to fine tune this research even further," Assoc Prof Al-Naji says.

Ensuring fairness of AI in healthcare requires cross-disciplinary collaboration

 Pursuing fair artificial intelligence (AI) for healthcare requires collaboration between experts across disciplines, says a global team of scientists led by Duke-NUS Medical School in a new perspective published in npj Digital Medicine.

While AI has demonstrated potential for healthcare insights, concerns around bias remain. "A fair model is expected to perform equally well across subgroups like age, gender and race. However, differences in performance may have underlying clinical reasons and may not necessarily indicate unfairness," explained first author Ms Liu Mingxuan, a PhD candidate in the Quantitative Biology and Medicine (Biostatistics & Health Data Science) Programme and Centre for Quantitative Medicine (CQM) at Duke-NUS.

"Focusing on equity -- that is, recognising factors like race, gender, etc., and adjusting the AI algorithm or its application to make sure more vulnerable groups get the care they need -- rather than complete equality, is likely a more reasonable approach for clinical AI," said Dr Ning Yilin, Research Fellow with CQM and a co-first-author of the paper. "Patient preferences and prognosis are also crucial considerations, as equal treatment does not always mean fair treatment. An example of this is age, which frequently factors into treatment decisions and outcomes."

The paper highlights key misalignments between AI fairness research and clinical needs. "Various metrics exist to measure model fairness, but choosing suitable ones for healthcare is difficult as they can conflict. Trade-offs are often inevitable," commented Associate Professor Liu Nan also from Duke-NUS' CQM, senior and corresponding author of the paper.

He added, "Differences detected between groups are frequently treated as biases to be mitigated in AI research. However, in the medical context, we must discern between meaningful differences and true biases requiring correction."

The authors emphasise the need to evaluate which attributes are considered 'sensitive' for each application. They say that actively engaging clinicians is vital for developing useful and fair AI models.

"Variables like race and ethnicity need careful handling as they may represent systemic biases or biological differences," said Assoc Prof Liu. "Clinicians can provide context, determine if differences are justified, and guide models towards equitable decisions."

Overall, the authors argue that pursuing fair AI for healthcare requires collaboration between experts in AI, medicine, ethics and beyond.

"Achieving fairness in the use of AI in healthcare is an important but highly complex issue. Despite extensive developments in fair AI methodologies, it remains challenging to translate them into actual clinical practice due to the nature of healthcare -- which involves biological, ethical and social considerations. In order to advance AI practices to benefit patient care, clinicians, AI and industry experts need to work together and take active steps towards addressing fairness in AI," said co-author Associate Professor Daniel Ting, Director of SingHealth's AI Office and Associate Professor from the SingHealth Duke-NUS Ophthalmology & Visual Sciences Academic Clinical Programme. He is also Senior Consultant at the Singapore National Eye Centre and Head of AI & Digital Innovation at the Singapore Eye Research Institute (SERI).

"This paper highlights the complexities of translating AI fairness techniques into ethical clinical applications. It represents our collective commitment to developing AI that augments clinicians with trustworthy insights to provide quality and equitable care enhanced by technology," remarked co-author Clinical Associate Professor Lionel Cheng Tim-Ee, Chief Data & Digital Officer, Clinical Director (AI) Future Health System Department, and Senior Consultant, Department of Diagnostic Radiology at Singapore General Hospital (SGH).

"Clinicians must be actively engaged in iterative communication with AI developers to ensure models align with medical ethics and context," stressed senior co-author Professor Marcus Ong, Director of the Health Services & Systems Research (HSSR) Programme at Duke-NUS, who is also Senior Consultant at SGH's Department of Emergency Medicine. "Good intentions alone cannot guarantee fair AI unless we have collective oversight from diverse experts, considering all social and ethical nuances. Pursuing equitable and unbiased AI to improve healthcare will require open, cross-disciplinary dialogues."

The perspective published in npj Digital Medicine represents an international collaboration between researchers from institutions across Singapore, Belgium, and the United States. Authors from across the SingHealth Duke-NUS Academic Medical Centre (including Duke-NUS, SingHealth, SGH, Singapore Eye Research Institute and Singapore National Eye Centre) worked together with experts from the University of Antwerp in Belgium as well as Weill Cornell Medicine, Massachusetts Institute of Technology, Beth Israel Deaconess Medical Center and Harvard T.H. Chan School of Public Health in the United States.

Professor Patrick Tan, Senior Vice-Dean for Research at Duke-NUS, commented, "This global cooperation exemplifies the cross-disciplinary dialogues required to advance fair AI techniques for enhancing healthcare. We hope this collaborative effort spanning Singapore, Europe, and the US provides valuable perspectives to inspire further multinational partnerships towards equitable and unbiased AI."

Scientists, philosophers identify nature's missing evolutionary law

 A paper in the journal Proceedings of the National Academy of Sciences today describes "a missing law of nature," recognizing for the first time an important norm within the natural world's workings.

In essence, the new law states that complex natural systems evolve to states of greater patterning, diversity, and complexity. In other words, evolution is not limited to life on Earth, it also occurs in other massively complex systems, from planets and stars to atoms, minerals, and more.

Authored by a nine-member team -- leading scientists from the Carnegie Institution for Science, the California Institute of Technology (Caltech) and Cornell University, and philosophers from the University of Colorado -- the work was funded by the John Templeton Foundation.

"Macroscopic" laws of nature describe and explain phenomena experienced daily in the natural world. Natural laws related to forces and motion, gravity, electromagnetism, and energy, for example, were described more than 150 years ago.

The new work presents a modern addition -- a macroscopic law recognizing evolution as a common feature of the natural world's complex systems, which are characterised as follows:

• They are formed from many different components, such as atoms, molecules, or cells, that can be arranged and rearranged repeatedly
• Are subject to natural processes that cause countless different arrangements to be formed
• Only a small fraction of all these configurations survive in a process called "selection for function."

Regardless of whether the system is living or nonliving, when a novel configuration works well and function improves, evolution occurs.

The authors' "Law of Increasing Functional Information" states that the system will evolve "if many different configurations of the system undergo selection for one or more functions."

"An important component of this proposed natural law is the idea of 'selection for function,'" says Carnegie astrobiologist Dr. Michael L. Wong, first author of the study.

In the case of biology, Darwin equated function primarily with survival -- the ability to live long enough to produce fertile offspring.

The new study expands that perspective, noting that at least three kinds of function occur in nature.

The most basic function is stability -- stable arrangements of atoms or molecules are selected to continue. Also chosen to persist are dynamic systems with ongoing supplies of energy.

The third and most interesting function is "novelty" -- the tendency of evolving systems to explore new configurations that sometimes lead to startling new behaviors or characteristics.

Life's evolutionary history is rich with novelties -- photosynthesis evolved when single cells learned to harness light energy, multicellular life evolved when cells learned to cooperate, and species evolved thanks to advantageous new behaviors such as swimming, walking, flying, and thinking.

The same sort of evolution happens in the mineral kingdom. The earliest minerals represent particularly stable arrangements of atoms. Those primordial minerals provided foundations for the next generations of minerals, which participated in life's origins. The evolution of life and minerals are intertwined, as life uses minerals for shells, teeth, and bones.

Indeed, Earth's minerals, which began with about 20 at the dawn of our Solar System, now number almost 6,000 known today thanks to ever more complex physical, chemical, and ultimately biological processes over 4.5 billion years.

In the case of stars, the paper notes that just two major elements -- hydrogen and helium -- formed the first stars shortly after the big bang. Those earliest stars used hydrogen and helium to make about 20 heavier chemical elements. And the next generation of stars built on that diversity to produce almost 100 more elements.

"Charles Darwin eloquently articulated the way plants and animals evolve by natural selection, with many variations and traits of individuals and many different configurations," says co-author Robert M. Hazen of Carnegie Science, a leader of the research.

"We contend that Darwinian theory is just a very special, very important case within a far larger natural phenomenon. The notion that selection for function drives evolution applies equally to stars, atoms, minerals, and many other conceptually equivalent situations where many configurations are subjected to selective pressure."

The co-authors themselves represent a unique multi-disciplinary configuration: three philosophers of science, two astrobiologists, a data scientist, a mineralogist, and a theoretical physicist.

Says Dr. Wong: "In this new paper, we consider evolution in the broadest sense -- change over time -- which subsumes Darwinian evolution based upon the particulars of 'descent with modification.'"

"The universe generates novel combinations of atoms, molecules, cells, etc. Those combinations that are stable and can go on to engender even more novelty will continue to evolve. This is what makes life the most striking example of evolution, but evolution is everywhere."

Among many implications, the paper offers:

1. Understanding into how differing systems possess varying degrees to which they can continue to evolve. "Potential complexity" or "future complexity" have been proposed as metrics of how much more complex an evolving system might become
2. Insights into how the rate of evolution of some systems can be influenced artificially. The notion of functional information suggests that the rate of evolution in a system might be increased in at least three ways: (1) by increasing the number and/or diversity of interacting agents, (2) by increasing the number of different configurations of the system; and/or 3) by enhancing the selective pressure on the system (for example, in chemical systems by more frequent cycles of heating/cooling or wetting/drying).
3. A deeper understanding of generative forces behind the creation and existence of complex phenomena in the universe, and the role of information in describing them
4. An understanding of life in the context of other complex evolving systems. Life shares certain conceptual equivalencies with other complex evolving systems, but the authors point to a future research direction, asking if there is something distinct about how life processes information on functionality.
5. Aiding the search for life elsewhere: if there is a demarcation between life and non-life that has to do with selection for function, can we identify the "rules of life" that allow us to discriminate that biotic dividing line in astrobiological investigations?
6. At a time when evolving AI systems are an increasing concern, a predictive law of information that characterizes how both natural and symbolic systems evolve is especially welcome

Laws of nature -- motion, gravity, electromagnetism, thermodynamics -- etc. codify the general behavior of various macroscopic natural systems across space and time.

The "law of increasing functional information" published today complements the 2nd law of thermodynamics, which states that the entropy (disorder) of an isolated system increases over time (and heat always flows from hotter to colder objects).

Grasping the three-dimensional morphology of kilonovae

 An advanced new three-dimensional (3D) computer simulation of the light emitted following a merger of two neutron stars has produced a similar sequence of spectroscopic features to an observed kilonova. "The unprecedented agreement between our simulations and the observation of kilonova AT2017gfo indicates that we understand broadly what has taken place in the explosion and aftermath," says Luke Shingles, scientist at GSI/FAIR and the leading author of the publication in The Astrophysical Journal Letters. Recent observations that combine both gravitational waves and visible light have pointed to neutron star mergers as the major site of this element production. The research was performed by scientists at GSI Helmholtzzentrum für Schwerionenforschung and Queen's University Belfast.

The interactions between electrons, ions, and photons within the material ejected from a neutron-star merger determine the light that we can see through telescopes. These processes and the emitted light can be modelled with computer simulations of radiative transfer. Researchers have recently produced, for the first time, a three-dimensional simulation that self-consistently follows the neutron-star merger, neutron-capture nucleosynthesis, energy deposited by radioactive decay, and radiative transfer with tens of millions of atomic transitions of heavy elements.

Being a 3D model, the observed light can be predicted for any viewing direction. When viewed nearly perpendicular to the orbital plane of the two neutron stars (as observational evidence indicates for the kilonova AT2017gfo) the model predicts a sequence of spectral distributions that look remarkably similar to what has been observed for AT2017gfo. "Research in this area will help us to understand the origins of elements heavier than iron (such as platinum and gold) that were mainly produced by the rapid neutron capture process in neutron star mergers," says Shingles.

About half of the elements heavier than iron are produced in an environment of extreme temperatures and neutron densities, as achieved when two neutron stars merge with each other. When they eventually spiral in toward each other and coalesce, the resulting explosion leads to the ejection of matter with the appropriate conditions to produce unstable neutron-rich heavy nuclei by a sequence of neutron captures and beta-decays. These nuclei decay to stability, liberating energy that powers an explosive 'kilonova' transient, a bright emission of light that rapidly fades in about a week.

The 3D simulation combines together several areas of physics, including the behavior of matter at high densities, the properties of unstable heavy nuclei, and atom-light interactions of heavy elements. Further challenges remain, such as accounting for the rate at which the spectral distribution changes, and the description of material ejected at late times. Future progress in this area will increase the precision with which we can predict and understand features in the spectra and will further our understanding of the conditions in which heavy elements were synthesized. A fundamental ingredient for these models is high quality atomic and nuclear experimental data as will be provided by the FAIR facility.

Pivotal breakthrough in adapting perovskite solar cells for renewable energy

 A huge step forward in the evolution of perovskite solar cells recorded by researchers at City University of Hong Kong (CityU) will have significant implications for renewable energy development.

The CityU innovation paves the way for commercialising perovskite solar cells, bringing us closer to an energy-efficient future powered by sustainable sources.

"The implications of this research are far-reaching, and its potential applications could revolutionise the solar energy industry," said Professor Zhu Zonglong of the Department of Chemistry at CityU, who collaborated with Professor Li Zhong'an at Huazhong University of Science and Technology.

New approach

Perovskite solar cells are a promising frontier in the solar energy landscape, known for their impressive power conversion efficiency. However, they have one significant drawback: thermal instability, i.e. they don't tend to perform well when exposed to high temperatures.

The team at CityU has engineered a unique type of self-assembled monolayer, or SAM for short, and anchored it on a nickel oxide surface as a charge extraction layer.

"Our approach has dramatically enhanced the thermal robustness of the cells," said Professor Zhu, adding that thermal stability is a significant barrier to the commercial deployment of perovskite solar cells.

action layer, our improved cells retain over 90% of their efficiency, boasting an impressive efficiency rate of 25.6%, even after operated under high temperatures, around (65 degrees Celsius) for over 1,000 hours. This is a milestone achievement," said Professor Zhu.

Raising the heat shield

The motivation for this research was born from a specific challenge in the solar energy sector: the thermal instability of perovskite solar cells.

"Despite their high power conversion efficiency, these solar cells are like a sports car that runs exceptionally well in cool weather but tends to overheat and underperform on a hot day. This was a significant roadblock preventing their widespread use," said Professor Zhu.

The CityU team has focused on the self-assembled monolayer (SAM), an essential part of these cells, and envisioned it as a heat-sensitive shield that needed reinforcement.

"We discovered that high-temperature exposure can cause the chemical bonds within SAM molecules to fracture, negatively impacting device performance. So our solution was akin to adding a heat-resistant armour -- a layer of nickel oxide nanoparticles, topped by a SAM, achieved through an integration of various experimental approaches and theoretical calculations," Professor Zhu said.

To counteract this issue, the CityU team introduced an innovative solution: anchoring the SAM onto an inherently stable nickel oxide surface, thereby enhancing the SAM's binding energy on the substrate. Also, they synthesised a new SAM molecule of their own, creating an innovative molecule that promotes more efficient charge extraction in perovskite devices.

Better efficiency in higher temperatures

The primary outcome of the research is the potential transformation of the solar energy landscape. By improving the thermal stability of perovskite solar cells through the innovatively designed SAMs, the team has laid the foundation for these cells to perform efficiently even in high-temperature conditions.

"This breakthrough is pivotal as it addresses a major obstacle that previously impeded wider adoption of perovskite solar cells. Our findings could significantly broaden the utilisation of these cells, pushing their application boundaries to environments and climates where high temperatures were a deterrent," said Professor Zhu.

The importance of these findings cannot be overstated. By bolstering the commercial viability of perovskite solar cells, CityU is not merely introducing a new player in the renewable energy market, it's setting the stage for a potential game-changer that could play a vital role in the global shift towards sustainable and energy-efficient sources.

"This technology, once fully commercialised, could help decrease our dependence on fossil fuels and contribute substantially to combating the global climate crisis," he added.