Category Archives: artificial intelligence

Americans Cautious About Advances in Artificial Intelligence and Human Enhancement Technologies

A new Pew Research Center survey finds that Americans believe that AI and human enhancements via technology have the potential to improve daily life and human abilities. Yet public views are also cautiously defined by 1) the context of how these technologies would be used, 2) what constraints would be in place and 3) who would stand to benefit – or lose – if these advances become widespread.

Public caution is mostly centered around concerns about privacy, autonomy, unintended consequences and the amount of change these developments might mean for humans and society.

This survey looks at a broad arc of scientific and technological developments – some in use now, some still emerging. Three highlight the burgeoning array of AI applications: the use of facial recognition technology by police, the use of algorithms by social media companies to find false information on their sites and the development of driverless passenger vehicles.

Another three are often described as types of human enhancements, revolve around developments tied to the convergence of AI, biotechnology, nanotechnology and other fields. They raise the possibility of dramatic changes to human abilities in the future: computer chip implants in the brain to advance people’s cognitive skills, gene editing to greatly reduce a baby’s risk of developing serious diseases or health conditions, and robotic exoskeletons with a built-in AI system to greatly increase strength for lifting in manual labor jobs.

Americans are far more positive than negative about the widespread use of facial recognition technology by police to monitor crowds and look for people who may have committed a crime: 46% of U.S. adults think this would be a good idea for society, while 27% think this would be a bad idea and another 27% are unsure.

By narrower margins, more describe the use of computer algorithms by social media companies to find false information on their sites as a good rather than bad idea for society (38% vs. 31%), and the pattern is similar for the use of robotic exoskeletons with a built-in AI system to increase strength for manual labor jobs (33% vs. 24%).

The survey of 10,260 U.S. adults was conducted between Nov. 1 and 7, 2021. There are five key themes that run through people’s answers on these questions.

  1. Americans’ judgments about the potential impact of this set of applications are varied and, for portions of the public, marked by uncertainty. [Link to graphic]
  2. Less than half of the public believes these technologies would improve things over the way they are now. [Link to graphic]
  3. Americans see a need for higher standards to assess the safety of technologies on the horizon than are currently used. [Link to graphic]
  4. There are sharp partisan divisions when people think about possible government regulation of these new and developing technologies.
  5. There are mitigating steps people say would make these AI and human enhancement developments more acceptable. [Link to graphic]

Source: Pew Research

AI and Human Enhancement: Americans’ Openness Is Tempered by a Range of Concerns

Mayo Clinic launches AI startup program to help early-stage companies get market-ready

Mayo Clinic launched its artificial intelligence startup program this week with four early-stage health tech companies. The program also features additional expert assistance from Google and electronic health records giant Epic.

This is a 20-week program called Mayo Clinic Platform_Accelerate. The aims are to help AI-based startups get market-ready by offering access to Mayo Clinic experts in regulatory, clinical, technology and business domains with a focus on AI model validation and clinical readiness.

According to John Halamka, M.D., president of the Mayo Clinic Platform, “Health tech startups are critical contributors to the cycle of innovation. We are excited to collaborate with these innovators to solve some of the most complex problems in medicine today.”

Mayo Clinic says its AI startup program is different from other typical accelerator programs. It focuses on helping drive a company’s success with valuable in-kind investments—from data sets to validation tools and from mentorship to clinical workflow planning, the company said on the program website.

As part of this in-kind investment, Mayo Clinic Platform will take an equity position in participating startups based on most recent valuations or a convertible note or SAFE, the organization said.

The four participating companies will work with data science experts to delineate AI model requirements, check for fairness and bias in their AI models and gain an understanding of the Food and Drug Administration clearance pathways.

The startups also will gain access to de-identified Mayo Clinic patient data, conduct model validation with guidance from data science experts, plan clinical validation studies such as clinical simulation or clinical trials and explore the potential to partner with Mayo.

Link: Mayo Clinic Platform_Accelerate

Predicting Early Antidepressant Outcomes For Children and Adolescents with MDD Using AI

Mayo Clinic researchers have taken the first step in using artificial intelligence (AI) to predict early outcomes with antidepressants in children and adolescents with major depressive disorder, in a study published in The Journal of Child Psychology and Psychiatry.

In the study, researchers identified variation in six depressive symptoms: difficulty having fun, social withdrawal, excessive fatigue, irritability, low self-esteem and depressed feelings.

They assessed these symptoms with the Children’s Depression Rating Scale-Revised to predict outcomes to 10 to 12 weeks of antidepressant pharmacotherapy:

  • The six symptoms predicted 10- to 12-week outcomes at four to six weeks in fluoxetine testing datasets, with an average accuracy of 73%.
  • The same six symptoms predicted 10- to 12-week outcomes at four to six weeks in duloxetine testing datasets, with an average accuracy of 76%.
  • In placebo-treated patients, predicting response and remission accuracy was significantly lower than for antidepressants at 67%.

According to the researchers, ‘these outcomes show the potential of AI and patient data to ensure children and adolescents receive treatment that has the highest likelihood of delivering therapeutic benefits with minimized side effects. We designed the algorithm to mimic a clinician’s logic of treatment management at an interim time point based on their estimated guess of whether a patient will likely or not benefit from pharmacotherapy at the current dose,” says Dr. Athreya, lead author. “Hence, it was essential for me as a computer engineer to embed and observe the practice closely to not only understand the needs of the patient, but also how AI can be consumed and useful to the clinician to benefit the patient.”

This preliminary work suggests that AI has promise for assisting clinical decisions by informing physicians on the selection, use and dosing of antidepressants for children and adolescents with major depressive disorder,” says Paul Croarkin, D.O., a Mayo Clinic psychiatrist and senior author of the study. “We saw improved predictions of treatment outcomes in samples of children and adolescents across two classes of antidepressants.”

Key points from the study

  • The optimal treatment of depression in children and adolescent is a substantial public health problem.
  • Machine learning and probabilistic graphical models were used to predict treatment outcomes with antidepressants in a training and testing databases.
  • Variation in six depression symptoms predicted outcomes with fluoxetine or duloxetine.
  • Future work should augment probabilistic graphical models with biological data to refine tools to assist decision making in clinical practice.

This work was a collaborative effort between the departments of Molecular Pharmacology and Experimental Therapeutics, and Psychiatry and Psychology, at Mayo Clinic, with support from Mayo Clinic’s Center for Individualized Medicine.

Study:
Athreya AP, Vande Voort JL, Shekunov J, et al. The Journal of Child Psychology and Psychiatry. 15 March 2022