We describe our system for the ArchEHR-QA Shared Task on answering clinical questions using electronic health records (EHRs). Our approach uses large language models in two steps: first, to find sentences in the EHR relevant to a clinician's question, and second, to generate a short, citation-supported response based on those sentences. We use few-shot prompting, self-consistency, and thresholding to improve the sentence classification step to decide which sentences are essential. We compare several models and find that a smaller 8B model performs better than a larger 70B model for identifying relevant information. Our results show that accurate sentence selection is critical for generating high-quality responses and that self-consistency with thresholding helps make these decisions more reliable.

What are the key questions radiologists should ask when assessing artificial intelligence (AI) vendors? While the list can be long, one important question is ascertaining the volume and nature of the data used to develop and train a given AI algorithm, according to Sonia Gupta, MD, an abdominal radiologist, and chief medical officer at Change Healthcare. In a recent video interview, Dr. Gupta said knowing the volume of cases that went into the training of an AI model is an important consideration as is the diversity of that data in terms of factors such as age, gender, health issues and comorbidities to name a few. "All of those factors will influence the model training and ultimately the performance of that AI algorithm," noted Dr. Gupta, who lectured about AI at the recent Society for Imaging Informatics in Medicine (SIIM) conference. "I encourage radiologists looking at potential AI vendors to dig into that information right off the bat."

According to statistics, AI projects fail at a high rate. With us, experience working with various clients has taught us that AI projects necessitate an entirely different strategy than normal mobile/web apps. This article defines the high-level method for effectively designing successful AI-powered applications. The International Data Corporation (IDC quotes as half of all Artificial Intelligence (AI) efforts fail. This is not the only accusation made by the IDC.

Wisdom is not algorithmic…You can't have an if this than that algorithm that actually equals wisdom. And if you can, than we are just an intermediate boot loader for the A.I's that are better creatures than us -- Daniel Smachtenberger Is first and foremost a marketplace. Now give this market self-learning A.I-driven tools, ever-increasing capacity for the aim of profit-making, and voila, here is where we are. Tech companies have become more powerful than nation-states. We are faced with a market dynamic that is unprecedented in human history.

"How can I solve my problem with AI?"- As Machine Learning and Artificial Intelligence reach more and more areas of daily life and enter all economic sectors, this question is often asked by decision makers eager to integrate AI into their business. While AI can offer great gains to businesses, in the following, you will see why jumping in with such a question is not an appropriate approach. Before diving into AI for your business problem, a well-defined business strategy must be established and the question of "Why should I use Machine Learning/Artificial Intelligence?" should be thoroughly considered. Being able to answer that question requires having the exact definition of the business problem: knowing the available data and desired output, having a plan for testing, monitoring and improving your solution, and being clear about the end use-case. After all, it's no use having a perfectly designed model from the data science team if you haven't planned how the rest of the company can use your model outputs.

If you're of a certain age and a fan of pop culture, the term "artificial intelligence" can't help but call up images of sentient creations that start off as man's loyal companions and servants, then somehow end up with the nuclear launch codes en route to wiping us off our own planet. The real-life incarnation of artificial intelligence is a 180-degree turn from its fictitious counterpart. The real disasters will come at the beginning of AI's service to your company if you haven't properly prepared for its incorporation. Your essential questions about AI, its functions and its capabilities must be answered before you choose a vendor and begin your transition. Otherwise, you're simply wasting time and manpower and will inevitably watch your competition sprint ahead into the future of your industry while you tread water and try not to drown.

It's not exactly news that Hugh Hefner, the perpetually robed Playboy founder who died Wednesday at the age of 91, is a polarizing figure. For decades Americans have disagreed about whether he should be remembered as a great liberator of Americans from their sexual puritanism or as a sexist exploiter of women. On Thursday, columnist Robin Abcarian came down strongly on the latter side, writing that although we shouldn't forget Hefner's support for smart journalism, reproductive rights and civil liberties, we should also not lose sight of the fact that his core business was the objectification of women -- mostly women under 30 -- and the exalting of exclusively male fantasies. Before Abcarian's column was published, the letters on Hefner's death reflected the typical mix of opinions we get after most notable celebrity passings: Several mentioned the existence of strong polarization over Hefner's work without taking a side, others reflected dispassionately on his work, and a few recounted their own experiences with Hefner. It was only in response to Abcarian's column that more readers started expressing stronger opinions on Hefner's work itself.