IBM is not the best fit for medical imaging and diagnosis in a variety of settings, according to new research by the software giant.
The Watson machine-learning tool, which uses artificial intelligence to analyse images of the human body, is far more capable at detecting cancerous tissue and organ systems than other medical imaging software currently in use, according a new study.
The company has been widely criticized for its failure to catch cancer and other disease, as well as for its expensive and poorly designed healthcare products.
The study, which surveyed more than 4,000 healthcare professionals across the US, Canada and Germany, found that Watson outperformed both commercial and academic healthcare software.
The IBM Watson system can see about 20% more images per second than commercial and commercial healthcare software, but only about 15% of those are useful to diagnose or treat cancer, the study found.
This is because Watson’s ability to recognize more complex images is based on its model of the body, rather than the anatomical features, said the study, published in the Journal of Medical Image Processing.
The model can take in thousands of images, and can be trained to recognize many different features, such as the size of the face, the shape of the nose and mouth, or even the presence of breathing tubes.
It is not clear if Watson can detect the structure of the lung or liver.
Other studies have found that it can spot more cancers than commercial software.
For example, IBM found that its Watson system could spot more lung cancers in people who have a genetic disease known as myelodysplastic syndrome (MDS).
However, it has been criticized for not being able to detect more diseases in people with existing medical conditions.
It also has been criticised for its high price tag.
It costs more than $100,000, but that is only because it takes the medical imaging technology developed by IBM and uses it.
The company has also been criticized over the lack of data and analysis that it offers.
“In general, we have found it to be less accurate in the detection of lung cancers than the commercial image analysis system,” the study authors wrote.
In the study’s conclusion, they said the Watson system was “not the most accurate tool available for detecting disease.”
In addition to this, the researchers said it was not able to spot cancerous tumours.
It would also take at least five hours for a single scan, and the software is not particularly well-suited for identifying a complex disease like cancer.
The researchers said that the new study’s findings are based on a number of other research papers published by IBM.
However, they noted that the study could not distinguish between “good” and “bad” results from IBM Watson.
“It’s difficult to distinguish between the two, as the results we see may be different from the results that other researchers are finding,” the researchers wrote.
However, they did note that Watson could also help with other medical tasks.
The report found that, in the case of cancer detection, the system could be used to detect the size and location of the cancerous lesion.
However the researchers also noted that there were other health care tasks where Watson’s software could be helpful, such in the assessment of blood glucose levels, and assessing the likelihood of a person being able continue their usual routine.
For healthcare professionals, the results could also provide some guidance.
“Although it is unclear whether Watson can provide a comprehensive set of diagnostic tools, the data from this study provides evidence of its potential in a number health-care settings, and shows that the Watson platform can provide important diagnostic capabilities,” the report said.
It was not immediately clear whether Watson would be able to help detect certain types of cancers.
However the researchers noted that they did not know if these were cancers that could be identified with existing software.