These algorithms might finish the scourge of tuberculosis


In some of the most remote and impoverished areas of the world, where respiratory illnesses abound and trained health professionals are afraid to kick, diagnosis is increasingly being powered by artificial intelligence and the internet.

In less than a minute, a new app on a phone or computer can scan an x-ray for signs of tuberculosis, Covid-19, and 27 other diseases.

TB, the deadliest infectious disease in the world, killed nearly 1.4 million people last year. The app called qXR is one of many A.I.-based tools that have been developed in recent years for screening and diagnosing TB.

The tools offer the hope of reporting the disease early and reducing the cost of unnecessary laboratory tests. On a large scale, they can also spot emerging disease clusters.

"Of all the applications by A.I., I think that digitally interpreting an image using an algorithm instead of a human radiologist is probably the most advanced," said Madhukar Pai, director of the McGill International TB Center in Montreal.

Artificial intelligence cannot replace clinicians, warned Dr. Pai and other experts. But the combination of A.I. and clinical expertise is proving powerful.

"The machine plus clinician is better than the clinician and also better than the machine alone," said Dr. Eric Topol, director of the Scripps Research Translational Institute in San Diego and author of a book on the use of A.I. In treatment.

In India, where roughly a quarter of the world's TB cases occur, there is an urgent need for an app that can be used to report the disease in remote locations.

Chinchpada Christian Hospital in Nandurbar, a small town in northwest India, serves members of the Bhil tribal community, some of whom travel up to 200 km to visit the center. The 50-bed hospital has eight doctors and only the rudimentary medical equipment.

Across the country, Simdega, one of the 20 poorest counties in India, is isolated from the nearest town of Rourkela by the nearly five-hour drive on bumpy roads. The tribal people in the district live in tiny hamlets surrounded by dense, evergreen forests. The Simdega medical center with 60 beds and three doctors is located in a clearing in the forest – "literally in the middle of nowhere," said Dr. George Mathew, the director.

The lean staff have to deal with whatever comes their way, "from malaria to myocardial infarction to cramps and head injuries," said Dr. Mathew. Over the years he has taught himself to read X-rays, and when at a loss, appeals to the radiologists among his distant friends and former colleagues.

Although the Nandurbar and Simdega are more than 800 miles apart, their populations are amazingly similar. Malaria, sickle cell disease, and tuberculosis are widespread among them, exacerbated by poverty, addiction to spiritual healers, and alcoholism – even among children.

"TB is often neglected and diagnosis is often delayed," said Dr. Ashita Singh, chief physician at Nandurbar Hospital. When people arrive at these medical centers, they are often "very, very sick and have not been examined anywhere else," she said.

In some patients, however, the X-rays carry signs that are too subtle for a non-specialist to see. "In this group of patients, A.I. technology can be of great benefit," said Dr. Singh.

The arrival of the coronavirus – and the lockdown that followed – separated these remote hospitals from nearby cities and also from radiologists. It also further delayed and complicated the diagnosis of TB, as both diseases affect the lungs.

A few months ago, both hospitals were using qXR, an app from the Indian company that was subsidized by the Indian government. The app allows the user to scan an X-ray. If it finds evidence of TB, it assigns the patient a risk assessment. Doctors can then do confirmatory tests on patients at greatest risk.

At the hospital in Nandurbar, the app helped diagnose TB in 20 patients in October, said Dr. Singh.

Apps like qXR can also be useful in places with low TB prevalence and for routine screening of people with H.I.V. who are at high risk of developing TB, as well as people with other conditions, experts say.

The coronavirus outbreak>

Things to know about testing

Confused by Coronavirus Testing Conditions? Let us help:

antibody: A protein produced by the immune system that can recognize and attach to certain types of viruses, bacteria or other invaders.Antibody test / serology test: A test that detects antibodies specific to the coronavirus. About a week after the coronavirus infects the body, antibodies start appearing in the blood. Because antibodies take so long to develop, an antibody test cannot reliably diagnose an ongoing infection. However, it can identify people who have been exposed to the coronavirus in the past.Antigen test: This test detects parts of coronavirus proteins called antigens. Antigen tests are quick and only take five minutes. However, they are less accurate than tests that detect genetic material from the virus.Coronavirus: Any virus that belongs to the Orthocoronavirinae virus family. The coronavirus that causes Covid-19 is known as SARS-CoV-2. Covid19: The disease caused by the new coronavirus. The name stands for Coronavirus Disease 2019.Isolation and quarantine: Isolation is separating people who know they have a contagious disease from those who are not sick. Quarantine refers to restricting the movement of people who have been exposed to a virus.Nasopharyngeal smear: A long, flexible rod with a soft swab that is inserted deep into the nose to collect samples from the space where the nasal cavity meets the throat. Samples for coronavirus tests can also be obtained with swabs that do not go as deep into the nose – sometimes called nasal swabs – or with mouth or throat swabs.Polymerase chain reaction (PCR): Scientists use PCR to make millions of copies of genetic material in a sample. With the help of PCR tests, researchers can detect the coronavirus even when it is scarce.Viral load: The amount of virus in a person's body. In people infected with the coronavirus, viral loads can peak before symptoms, if any.

"Most chest x-rays for people suspected of having tuberculosis are read by people who are unable to interpret them remotely," said Dr. Richard E. Chaisson, TB expert at Johns Hopkins University. "If there is an A.I. A package that could read the X-ray and CT scans for you in a remote emergency room would be a big step forward. "

qXR is one of the more promising A.I.-based apps for detecting TB. The company that developed the app didn't realize this potential until it was suggested by a doctor at an Indian hospital a few years ago.

In studies in which various A.I. Applications carried out by the Stop TB Partnership, all A.I. Apps outperformed experienced human readers, and qXR seemed to do the best.

According to Prashant Warier, CEO of, the app identifies TB with an accuracy of 95 percent. However, this level of precision is not based on real-world conditions that Dr. Topol named a "common problem" with A.I.-based apps. A TB program might be less accurate in the US or Western Europe than in India, as the prevalence of the disease is lower in those places, added Dr. Topol added.

The app was only tested on adults but is now used on children 6 years and older. Chest x-rays are especially useful in pediatric TB because about 70 percent of cases in children cannot be confirmed by laboratory tests, said Dr. Silvia S. Chiang, a specialist in pediatric TB at Brown University.

"There is a huge shortage of trained professionals who are comfortable interpreting pediatric chest x-rays," she said. "Developing and validating computerized x-ray reading technologies in children would therefore be very helpful." said it is testing its app on children in Bangladesh and will release the data early next year. In the meantime, qXR and other apps will keep improving as they learn over time.

"The more x-rays you give the animal, the better it gets," said Dr. Pai.

The experts were optimistic that A.I.-based apps could have a huge impact on the fight against TB, especially in countries such as India, where medical resources are lacking.

"I only dream of a time when something like this will be available to all small primary and secondary health centers in the government sector that are reluctant to take x-rays because they don't have the confidence to read them." Said Dr. Singh. "If this were made available to every X-ray center in rural India, we could beat TB."