4
Luckily, Cerner is far from the only company using AWS machine learning
capabilities to develop healthcare models. Developing speed and accuracy in
machine learning models is an exciting application of the technology.
Think again of that elderly patient who was waiting for the results of his CT
scan. Smart radiology start-up Aidoc would accelerate the review process. The
company's deep learning algorithm draws on knowledge from millions of past
records to accurately identify the growing embolism, then immediately places a
red flag on the patient's images in the ED radiology queue. Rather than sifting
through the scans in chronological order, the anomaly is pushed to the top and
the patient is rushed to critical care.
"Everything improves if you get to that patient in time," says Aidoc CEO
Elad Wallach. Now deployed in hundreds of facilities worldwide, the Aidoc
technology has already seen measurable results: At a major East Coast medical
center, machine learning applied to CT images reduced lengths of stay by three-
fourths of a day, and reduced time spent in the ED by 59 minutes. A similar
prospective study reduced the ER turnaround time of CT scans for intracranial
hemorrhages from 53 minutes to 46 minutes.
In addition to being life-saving for patients, this kind of support is sorely
needed by doctors as well. Limited staff in radiology departments in particular
can cause strain, but some 65% of doctors across disciplines report feeling
overworked. Machine learning tools can alert doctors to anomalies in scans and
flag possible diagnoses for doctors to review, saving doctors valuable time in
the review process.
Making Decisions When it Matters:
Improving Diagnosis
Everything
improves
if you get
to that
patient
in time."
Elad Wallach,
Aidoc CEO
"
