Artificial intelligence (AI) is a vast field that is developing fast. Its possible applications can be confusing and intimidating for patients. At the same time, we face challenges and needs in healthcare where AI can provide support and offers solutions. Emily Lewis, a digital health innovator, knows both the industry’s and the patient’s perspective and will shed light on different aspects of the topic. She is confident that we can use technology to re-humanize healthcare and bring connection.

In the first part of our series about “AI in Healthcare”, Emily explains how the use of AI can help socially challenged populations (e.g. people diagnosed with autism spectrum disorder or psychiatric disorders) or seniors. Here’s how:

People with Autism Spectrum Disorder (ASD)

Early Detection and Intervention

AI systems can analyze patterns in children’s behavior to detect early signs of ASD. For example, research has shown that AI algorithms can be used to analyze a child’s use of language. For instance, an algorithm could detect unusual patterns or delays in language development. Additionally, some researchers are applying AI and computer vision techniques to analyze facial expressions and social interactions.

• CanvasDx: Cognoa has developed a tool called CanvasDx for assessing the risk of Autism Spectrum Disorder (ASD) in children. The tool utilizes machine learning to analyze parent-provided home videos and data about a child’s behavior and development. The AI scoring of this data is similar to expert clinicians in distinguishing children with ASD. The software then generates a report indicating whether the child is at risk for ASD, which can aid healthcare providers in making an earlier diagnosis.
• Early Autism Detection App: A team at Duke University has developed an app that uses machine learning to analyze eye movements and emotional responses of children while they watch videos on a smartphone or tablet. This app has shown potential in detecting signs of autism in children as young as 6 months old.

Social Skills Training

There are AI-based applications that use virtual reality and computer-based interaction to help individuals with autism improve their social skills. Moreover, AI-powered robots can be used to teach basic social interactions like maintaining eye contact, recognizing facial expressions, and understanding tone of voice. These solutions use AI to personalize the learning experience, simulate real-life social interactions, and help individuals improve their social and cognitive skills.

• Robots: Social robots like Kaspar, Milo, and NAO have been used to teach social skills to children with autism. Although these toys are not cheap, they use AI to simulate social interactions and provide a controlled environment for the child to learn and practice skills such as turn-taking, eye contact, and emotional recognition.
• Apps: Several apps use AI to customize learning for people with autism. For example, the app “Otsimo” provides AI-powered educational games designed according to the principles of Applied Behavior Analysis (ABA) therapy. The app’s AI learns from the child’s interactions and adjusts the difficulty level and content accordingly.
• Virtual Reality: AI can be integrated with Virtual Reality (VR) to provide immersive social skills training. For example, the software “Floreo” uses VR to teach social skills and coping strategies in a fun, engaging environment. The software uses AI to adapt the virtual lessons based on the individual’s performance. Moreover, the SocialWiseVR app provides a suite of interactive immersive experiences whereby those with autism can experience first-hand a variety of social interactions, within the safety of the therapist environment. Example scenarios include going on a first date, being at a party, and dealing with authority figures. The solution uses a scaffold learning format with success gates applied to each scenario and a gamified interface to incentivize and reward learner’s progress.
• AI Tutoring Systems: Systems like “The Social Express” and “Brain Power System” use AI to teach social norms and behaviors. These systems use video modeling and AI to assess the user’s understanding and adjust the teaching approach as needed.

Support for Day-to-Day Tasks

AI can provide personalized recommendations and reminders for people with autism to help them manage their daily routines and reduce anxiety.

People with Psychiatric Disorders

Early Warning Sign Detection

AI can monitor online activity and identify potential warning signs of a psychiatric crisis. For example, changes in social media activity or search engine use can indicate increased risk.

• Ellipsis Health: This company uses AI to analyze natural speech to assess patients’ mental health. Their technology can detect signs of anxiety and depression, allowing for earlier and more proactive treatment.
• Moodpath: This app asks daily questions to users to assess their wellbeing and screen for symptoms of depression. The AI uses these responses to generate a bi-weekly mental health assessment that users can share with healthcare professionals.

Symptom Management

Machine learning can analyze user data to predict and manage symptoms. For instance, apps can track a user’s sleep, activity, and self-reported mood to predict depressive episodes or anxiety spikes.

• Woebot: This AI-driven chatbot uses principles of cognitive-behavioral therapy (CBT) to help users manage symptoms of depression, anxiety, and other mental health conditions. Woebot checks in with users daily and provides helpful tools and insights based on the user’s mood and needs.
• Youper: An AI-powered emotional health assistant app that uses cognitive-behavioral techniques, mindfulness, and other evidence-based approaches to help users monitor and manage their emotional health. Youper allows users to track their moods, discover triggers, and improve mental health over time.

Therapy

AI-powered chatbots can provide support between therapy sessions. These chatbots can offer cognitive behavioral therapy (CBT) strategies, relaxation exercises, and even crisis intervention resources.

• Ginger: This is a complete behavioral health system combining human coaches, therapists, and AI chatbots. It provides personalized self-care activities and tracks users’ mood and behavior to offer insights and real-time care.
• Talkspace: While primarily a platform for online therapy with licensed professionals, it also incorporates AI elements to match users with the best fitting therapists based on their needs.

Seniors

Maintaining Independence

AI can assist seniors in managing their medication, scheduling appointments, and even completing chores through voice-activated assistants or robotics.

• Medication Management: AI-based medication management systems like MedMinder and Hero can remind seniors to take their medications on time, dispense the correct dosage, and alert caregivers or family members if a dose is missed.

Ensuring Safety

AI can monitor seniors’ activities to detect irregularities or emergencies. For example, smart home technologies can detect falls or changes in routine that may indicate health issues. Wearable devices can monitor vital signs and alert medical professionals if necessary.

Social Interaction

AI-powered companions can provide conversation, remind seniors of appointments or tasks, and even detect mood or cognitive changes that might indicate a need for intervention.

• AI Companions: Robotic companions such as ElliQ and Mabu provide social interaction and can also offer reminders for medication, appointments, and exercise.

Conclusion

It’s important to note that while AI holds great potential in these areas, there are also important considerations around privacy, accuracy, and the risk of over-reliance on technology. The technology is not a substitute for human care, but rather a tool that can supplement and enhance the care provided by humans.