1 . Artificial intelligence is almost twice as accurate as a biopsy (活组织检查) at judging the aggressiveness of some cancers, experts say. Cancer kills 10 million people globally every year, according to the WHO. But for patients the disease can be prevented if detected instantly and dealt with quickly.

A recent study suggested an AI algorithm (算法) was far better than a biopsy at correctly grading the aggressiveness of sarcomas (肉瘤), a rare form of cancer. Researchers hope AI will improve outcomes for patients by giving doctors a more accurate way of grading tumours (肿瘤). Because high-grade tumours can indicate aggressive disease, the tool could help ensure those high-risk patients are identified more quickly and treated instantly. Low-risk patients could also be spared unnecessary treatments, follow-up scans and hospital visits.

Researchers say the algorithm could be applied to other types of cancer in future. The team specifically looked at retroperitoneal sarcomas, which develop at the back of the abdomen and are difficult to diagnose (诊断) and treat due to their location. They used CT scans from 170 patients with the two most common forms of retroperitoneal sarcoma — leiomyosarcoma and liposarcoma. Using data from these scans they created an AI algorithm, which was then tested on 89 patients in other countries. In grading how aggressive the tumour was, the technology was accurate in 82% of the cases, while biopsies were 44%.

AI could also recognize leiomyosarcoma and liposarcoma in 84% of sarcomas tested, while radiologists were able to identify them in 65% of the cases. Christina Messiou, the study leader, said: “We’re incredibly excited by the potential of this state-of-the-art technology, which could lead to patients having better outcomes through faster diagnosis. As patients with retroperitoneal sarcoma are routinely scanned with CT, we hope this tool will eventually be used globally, ensuring that not just specialist centres can reliably identify and grade the disease.”

Richard Davison, chief executive of Sarcoma UK, said the results looked “very promising”. He added: “People are more likely to survive sarcoma if diagnosed early. One in six people with sarcoma cancer wait more than a year to receive an accurate diagnosis, so any research that helps patients receive better treatment and support is welcome.”

2 . “Your mother needs a new heart,” my father told me when I called on that December afternoon. An unrelenting optimist, he spoke as if she merely needed to have a part replaced. But, although my two sisters and I knew that our mother had heart problems, this news still made us frozen for a while with our eyes widening in disbelief.

Dr. Marc Semigran of the transplant team reviewed my mother’s medical history. She’d had an irregular and rapid heartbeat for most of her life. Her present treatment — the use of a series of cardioversions, or electric jolts, to restore a normal heartbeat — would not work in the long-term. She had an enlarged and weakened heart, as well as a faulty valve.

“With medication,” Dr. Semigran said, “you have a 60 percent chance of living six months. You could have a longer life with a transplant, but there are risks. You’re at the top end of the age group of sixty years old. The lungs and other organs must be healthy and strong. While the transplant surgery is actually a straightforward procedure, acceptance by the body is the difficult thing.”

My family came together, trying to provide strength and work out what to do. We’d already gone from shock, over our mother’s condition, to worry that she wouldn’t be a suitable recipient. Despite of the risk, we chose to believe that she would make it eventually.

Word came later in December that she had been accepted into the programme. Dr. Jeremy Ruskin told us one of the reasons she had been accepted was that she had such strong family support.

One Monday in May, at about 8 p. m., my mother received a phone call from the hospital that a heart was available. As she was about to be wheeled off, my father took her face in his hands and looked into her eyes. His look said everything about their 42-year relationship.

The heart transplant operation was successful and the conditions could not have been better.

The irony of the transplant process is that one family’s loss is another’s gain; that tragedy begets fortune. It is a kind of life after death, our hearts beating beyond us. We developed a feeling of love for this new part, of gratitude for the doctors, for the process, and for those people who made a decision just for humanity.

3 . Clinical intuition is finally getting the respect it may have long deserved. New research is supporting a process that combines rapid judgments and perceptions that occur outside of conscious awareness — a way of knowing something without knowing how you know it. The value of clinical intuition in medicine is now being studied and becoming particularly relevant as interest grows in developing AI systems that can analyze medical data to diagnose or treat patients.

A study in 2023 found that clinical intuition from physiotherapists on the prognosis for functional recovery among patients was closely associated with the recovery afterwards. Another study concluded that in comparison to relying only on clinical data, “preoperative surgeon intuition alone is an independent predictor of patient outcomes.”

“This is a cognitive process. Clinical intuition is about expertise, knowledge, and pattern recognition that accumulate through experience. The mind is putting together all kinds of information and sequencing it in ways that say this person is really sick — or not,” says Meredith Vanstone, an associate professor in McMaster University. Given the depth of this cognitive process, some experts are skeptical that AI technology can make medical decisions as well as a human physician.

“As physicians go through years of interactions with patients and seeing thousands of cases, those gut feelings become a sort of summary statement of all the experiences they’ve seen,” says Mohammad Ghassemi, a researcher at Michigan State University. “Clinicians can observe different things that are not always captured or made available by machines.”

Researchers found that physicians and nurses practicing in a medical specialty involving a high likelihood of an emergency or dimensions of complexity are more likely to use intuitive decision-making in their practice. “As surgeons, the first thing we need to do is look at the patient because the structured data may not be consistent with what we see with the patient,” says Gabriel Brat from Harvard Medical School.

Research has found that clinical intuition from nurses and physicians about the chances that older patients visiting emergency departments would die or have other adverse outcomes within 30 days was highly accurate.

None of known experiences suggest that clinical intuition should be blindly acted upon. Making clinical decisions solely on the basis of a physician’s intuition isn’t the way to go, experts say. But relying solely on medical algorithms (mathematical models) that generate predictions about how a patient is likely to respond to different treatments isn’t the best course of action either.

To optimize patient care, some experts believe a hybrid approach that integrates clinical intuition, predictive algorithms, patient preferences, and other key factors is essential.

Over the years, acupuncture (针灸)     1     (become) a global therapy. It has seen many advancements in scientific research and modern medicine. It is now a popular treatment for a variety of     2     (disease). According to a 2023 WHO report, acupuncture has been used in 113 member countries,     3     shows its widespread recognition and application.

5 . A quick increase of dopamine (多巴胺) shifts mice into a dreamy stage of sleep. In the mice’s brains, the chemical messenger triggers rapid-eye-movement sleep, or REM, researchers report in the March 4 Science.

These new results are some of the first to show a trigger for the shifts. Understanding these transitions in more detail could ultimately point to ways to treat sleep disorders in people.

Certain nerve cells in the ventral tegmental area of the mouse brain can pump out dopamine, a molecule that has been linked to pleasure, movement and learning, which is then delivered dopamine to the amygdalae, two almond-shaped structures deep in the brain that are closely tied to emotions.

Using a molecular sensor that can tell exactly when and where dopamine is released, the researchers saw that dopamine levels rose in the amygdalae just before mice shifted from non-REM sleep to REM sleep.

Next, the researchers forced the mice into the REM phase by controlling those dopamine-producing nerve cells using lasers and genetic techniques. Compelled with light, the nerve cells released dopamine in the amygdalae while mice were in non-REM sleep. The mice then shifted into REM sleep sooner than they typically did, after an average of about two minutes compared with about eight minutes for mice that weren’t prompted to release dopamine. Stimulating these cells every half hour increased the mice’s total amount of REM sleep.

Additional experiments suggest that these dopamine-making nerve cells may also be involved in aspects of narcolepsy (嗜睡症). A sudden loss of muscle tone, called cataplexy, shares features with REM sleep and can accompany narcolepsy. Stimulating these dopamine-making nerve cells while mice were awake caused the mice to stop moving and fall directly into REM sleep.

The results help clarify a trigger for REM in mice; whether a similar thing happens in people isn’t known. Earlier studies have found that nerve cells in people’s amygdalae are active during REM sleep.

Many questions remain. Drugs that change dopamine levels in people don’t seem to have big effects on REM sleep and cataplexy. But these drugs affect the whole brain, and it’s possible that they are just not selective enough.

8 . Digital Medicine

Could the next suggestion from your doctor be downloading an app? Collectively known as digital medicine, a large number of apps in use or under development can now detect or monitor mental and physical disorders or directly administer therapies.     1    

Many detection aids rely on mobile devices to record such features as users’ voices, locations, facial expressions, exercise and sleep; then they apply artificial intelligence to mark the possible change of a condition. Some smart watches, for instance, contain a sensor that automatically detects and warns people of a dangerous heart rate.     2     These wearable detection aids will not replace a doctor any time soon but can be helpful partners in stressing concerns that need follow-up.     3     Some are being developed to detect things such as cancerous DNA, stomach bleeds, body temperature and oxygen levels. The sensors inside your body can then send the data to apps for recording.

    4     For the most part, those intended to diagnose or treat disorders must be proved safe and effective in clinical trials and earn regulatory approval; some may need a doctor’s permission.

Clearly, society must move into the future of digital medicine with care — ensuring that the apps go through strict testing, protect privacy and go smoothly into doctors’ work. With such protections in place, healthcare costs could be saved by marking unhealthy behaviours and helping people to make changes before diseases set in.     5     For researchers, the patterns that emerge will provide them with novel ideas for how best to build healthier habits and prevent diseases.