Artificial intelligence (AI), healthcare, and investing in stocks are three of my favorite topics. So when you combine all three of them, I really get excited. 

There are quite a few great stocks of companies that are harnessing the power of AI in improving healthcare. IBM (NYSE:IBM), for example, is already changing healthcare in several ways with its Watson AI technology.  NVIDIA (NASDAQ:NVDA) developed an AI platform that's designed specifically for healthcare applications.

I like IBM and NVIDIA. But if you're looking for the best healthcare AI stock on the market right now, check out Google parent Alphabet (NASDAQ:GOOG) (NASDAQ:GOOGL).

From a transparent screen filled with adjacent hexagonal icons,a man touches an icon labeled AI.

Image source: Getty Images.

Follow the money

Where do most healthcare dollars go in the U.S.? It's not prescription drugs. It's not even costly long-term care. Nearly one-third of national health expenditures in the U.S. come from hospital care, according to the Centers for Disease Control and Prevention. So if you want to use AI to really make a difference in healthcare, the most practical area to focus on would be in reducing hospital costs.

That's exactly what Alphabet is doing. Researchers from the company's Google business unit teamed up with others from the University of California, Stanford University, and the University of Chicago to develop an AI system that could predict the outcomes of hospital visits.

In January, the Google-led team published a paper detailing their effort to use AI in predictive modeling of hospital visits. The researchers used electronic health record data from two U.S. academic medical centers with 216,221 adult patients. This data included nearly 46.9 billion data points.

The results from the AI initiative were impressive. Deep-learning neural networks were able to predict in-hospital mortality, 30-day unplanned readmissions, prolonged length of stay, and all of a patient's final diagnoses with high levels of accuracy that were significantly more accurate than current models.  

Why would predicting these outcomes matter? Consider that of the $26 billion spent each year on hospital readmissions, an estimated $17 billion is for readmissions that could have been prevented if patients received appropriate care on their first visits. And any reduction in length of stay has the potential to dramatically lower costs. 

A great vision

Google's healthcare AI team thinks "that AI is poised to transform medicine, delivering new, assistive technologies that will empower doctors to better serve their patients." One way this vision is being achieved is by using AI to help with early diagnosis of diabetic retinopathy.

Closeup of eye with superimposed computer image

Image source: Getty Images.

Diabetic retinopathy results from high blood sugar levels causing damage to blood vessels in the retina. These damaged blood vessels can swell, leak blood, or even close altogether. Any of these conditions can cause loss of sight. While diabetic retinopathy is the fastest-growing cause of blindness in the world, it's also preventable. 

Retinal scans can detect diabetic retinopathy. If found early, the disease can be treated. But if not, patients can lose part or all of their eyesight. The problem is that there aren't enough trained eye doctors in the world to do the screening required to detect diabetic retinopathy early enough.

That's where Alphabet's AI could make a difference. The company worked with doctors in the U.S. and in India to create a dataset of 128,000 images of retinal scans. These images were reviewed by a board of ophthalmologists to determine which images showed signs of diabetic retinopathy. Alphabet's AI system was then trained on a subset of the data. The system ultimately was able to detect the disease slightly better than human ophthalmologists.

There's more work to do, though. These results used only 2D images, while ophthalmologists also often use 3D images to diagnose diabetic retinopathy. Alphabet's DeepMind team is hard at work applying AI to make diagnoses using 3D images.

Taking stock

As I mentioned, Alphabet isn't the only company doing some really cool things in using AI in healthcare. But I think it's the best stock of all of them, for a couple of reasons.

First, unlike IBM, Alphabet's revenue and earnings are growing nicely. Although IBM did report year-over-year revenue growth in the fourth quarter, that followed 22 straight quarters of year-over-year revenue declines. Alphabet, on the other hand, has seen its trailing-12-month revenue jump nearly 130% over the last five years.

Second, unlike NVIDIA, Alphabet stock is attractively valued. NVIDIA shares trade at more than 32 times expected earnings, while Alphabet's forward earnings multiple stands below 22. With growth prospects factored in, Alphabet stock is an even better bargain than NVIDIA.

In my view, the opportunities are tremendous for any company that makes significant progress in applying AI to healthcare. Alphabet fits that category and has an appealing valuation and financial growth story to boot. That should be enough to get any investor excited.  

This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.