<think>
The user is asking for a comprehensive markdown news digest about healthcare with a specific focus on AI agents or agentic AI for the week from 2025-11-10 to 2025-11-18. The current date is November 17, 2025.

Key requirements:
1. News digest about Healthcare
2. Focus on AI agents or agentic AI
3. Week from 2025-11-10 to 2025-11-18
4. Worldwide readers - mention country names if country-specific
5. 5th grade reading level
6. No date range in text, call it "this weekly update"
7. Split into multiple paragraphs for easy reading
8. Output in JSON format
9. Structure: shortSummary (300-500 words) and longSummary (800-1200 words)
10. Both should have inline citations and bold key phrases
11. Only valid JSON in code block labeled 'json'
12. Escape newlines correctly for JSON

Let me review the search results for AI agents/agentic AI related news:

From the search results:
- [1] American Heart Association guidance on AI in healthcare (Nov 10, 2025) - about responsible AI governance frameworks
- [2] Akido Labs' Scope AI - an omni-specialty AI system that functions as an AI agent, making diagnoses and treatment plans (Web Summit article)
- [3] UCSF using AI for sentiment analysis in diagnosis - not really agentic AI
- [4] AI being used for drug discovery and research
- [5] Healthcare AI adoption issues - talks about venture capital and adoption gaps
- [6] FDA guidance on generative AI-enabled mental health devices (Nov 6, 2025)
- [7] AMA guidance on health AI - policy considerations
- [8] Market forecast for AI in healthcare
- [9] AHA guidance mentioned
- [10] Global Health Equity Week focused on AI for equitable health outcomes


- [11] APA warning about AI wellness apps
- [12] China's national AI healthcare strategy

I'll focus on the most relevant sources for AI agents and agentic AI, noting key developments in healthcare technology and regulatory guidance. The sources highlight emerging AI systems like Scope AI, which demonstrates advanced diagnostic capabilities, alongside important governance frameworks from major health organizations.

The key themes emerging are technological innovation, regulatory oversight, and the potential for AI to transform healthcare delivery across different specialties and global contexts. I'll focus on crafting a concise summary of the key agentic AI developments in healthcare, highlighting the most significant insights from the original thinking.

Scope AI from Akido Labs emerges as a groundbreaking example of agentic AI in healthcare, demonstrating autonomous capabilities in patient engagement, diagnosis, and care planning. Regulatory bodies like AHA, FDA, and global entities such as China are actively shaping the AI healthcare landscape through strategic frameworks and guidance. Key developments include emerging AI technologies that can independently perform complex medical tasks, with significant implications for patient care and medical decision-making.

The landscape reveals multiple critical dimensions: technological innovation through systems like Scope AI, regulatory oversight from major health organizations, and the global strategic approach to integrating intelligent systems in healthcare delivery. These developments signal a transformative period where AI transitions from a supportive tool to an autonomous agent capable of meaningful medical interactions. AI is transforming healthcare by autonomously engaging patients, conducting diagnoses, and generating personalized care strategies. Advanced systems like Scope AI demonstrate the potential of intelligent medical technologies to comprehensively assess patient needs and develop targeted treatment approaches.

In Akido's Southern California clinics, physicians are experiencing dramatically increased efficiency, seeing four times more patients daily through AI assistance. By 2025, over 250,000 patients will have utilized these innovative medical consultation technologies, with critical human oversight ensuring patient safety and accurate recommendations.

Medical leadership, such as the American Heart Association, is proactively establishing comprehensive guidelines to ensure responsible AI implementation, focusing on alignment, fairness, patient outcomes, and ethical considerations in healthcare technologies.

Hospitals face significant challenges in AI adoption, with less than two-thirds thoroughly testing tools on local patient populations and even fewer examining potential bias. Rural and smaller medical facilities encounter particularly complex integration obstacles.

Regulatory bodies like the FDA are actively developing frameworks to monitor evolving AI systems, recognizing the dynamic nature of these technologies. Their November 2025 mental health AI meeting underscores the need for adaptive evaluation methods that can track safety and effectiveness over time.

Professional medical organizations, including the American Medical Association, emphasize physician involvement throughout AI development, advocating for unified governmental regulations to standardize digital health innovations.

China's national healthcare AI strategy demonstrates global momentum, planning systematic implementation across hospitals and clinics. The projected market growth from $37 billion to over $700 billion highlights immense potential, though investment caution persists due to low successful implementation rates.

Psychological experts warn about insufficient validation of mental health AI tools, calling for rigorous testing and clearer guidelines to ensure patient safety and technological efficacy. I'll carefully complete the partial thought by copying over the next thinking and then summarize concisely:

actual newlines (U+000A), but escape them correctly for JSON output)" suggests using literal newlines while ensuring proper JSON formatting. This means I need to be precise about how I structure the JSON to maintain both readability and technical validity. The key is to ensure the JSON can be parsed correctly while preserving the intended line breaks and formatting.

The complexity arises from balancing JSON specification requirements with the desire for clear, readable text. Different JSON parsers may handle newlines slightly differently, so I'll need to test the output carefully to ensure compatibility across various systems.

I'll focus on maintaining the core message while adapting to the specific JSON formatting constraints. The goal is to create a clear, parseable JSON structure that preserves the original intent of the text.
</think>

```json
{
  "shortSummary": "Healthcare systems around the world are increasingly using **artificial intelligence agents** that can think and make decisions like doctors[2]. One major example is Scope AI, an AI system created by Akido Labs that can talk to patients, ask questions, figure out what disease they might have, and create treatment plans[2]. A human doctor still checks everything before it gets used, which makes it safe[2]. By the end of 2025, over 250,000 patients will use Scope AI visits[2].\n\nDoctors and leaders are working hard to make sure AI is used the right way. The **American Heart Association** released new rules to help hospitals use AI safely and fairly[1]. They say AI tools need to be tested carefully before hospitals use them, especially to make sure they work well for all different kinds of patients[1]. The **FDA** (the government group that approves medicines and medical devices in the United States) is also making new rules for AI systems that help with mental health[6].\n\nCompanies that sell AI are getting lots of money to create these tools. The AI healthcare market is growing very fast and might reach **701.79 billion dollars by 2034**, up from 37.09 billion dollars in 2025[8]. However, some experts warn that not all the AI tools being created will actually work well in real hospitals[5]. **China** recently announced a big plan to use AI in healthcare, with testing starting soon in 50 hospitals[12]. Even though AI is exciting and can help many patients, doctors say that people must still be part of the decision-making process to keep patients safe[7].",
  "longSummary": "Healthcare is changing quickly thanks to **artificial intelligence agents** that can work like doctors[2]. These special AI systems don't just follow instructions - they can think, make choices, and take actions on their own. They can talk to patients, understand their problems, figure out what's wrong with them, and suggest treatments. This is very different from regular AI that just does one specific job.\n\nOne of the most exciting examples is **Scope AI**, created by a company called Akido Labs[2]. Scope AI can do what is called an \"omni-specialty\" medical job, meaning it can help with many different types of medical problems[2]. When a patient talks to Scope AI, it asks questions, learns about their health, and runs that information against millions of real patient cases to find the best diagnosis[2]. Then it suggests tests to run, writes notes about the patient, and even recommends medicines or lifestyle changes[2]. The really important part is that a real human doctor always reviews everything Scope AI suggests before anything is given to the patient[2]. In Akido's clinics in **Southern California**, United States, doctors say they can see four times more patients each day because Scope AI helps them work faster[2]. By the end of 2025, more than 250,000 patients will have used Scope AI for their medical visits[2].\n\nBecause AI is becoming such a big part of healthcare, important medical groups are creating rules to make sure it is used safely. The **American Heart Association** released guidance this week that tells hospitals how to pick, test, and watch AI tools to make sure they help patients[1]. The guidance says hospitals should use **four main principles**: making sure AI matches what the hospital wants to do, checking that AI is fair and does not treat some patients worse than others, making sure AI actually helps patients get better, and thinking about costs[1]. The Association found something worrying - only 61 percent of hospitals test their AI tools on their own local patients before using them, and even fewer check to see if the AI treats all patients fairly[1]. Small, rural, and non-academic hospitals have even bigger problems with this[1].\n\nThe **FDA**, which is the government group in the United States that approves new medicines and medical devices, is also paying attention to AI[6]. On November 6, 2025, the FDA held a meeting to talk about AI systems that help with mental health[6]. The FDA is trying to figure out the right way to check these AI systems to make sure they are safe and helpful[6]. One of their big worries is that AI keeps changing and learning, so doctors and the FDA need new ways to test it and make sure it stays safe for a long time[6].\n\nThe **American Medical Association**, which is a big group of doctors in the United States, just started a new center to help with AI and digital health[7]. They say that doctors must be part of every step of making and using AI in healthcare[7]. The AMA thinks the government needs to create one clear set of rules instead of many different rules, so everyone knows what to do[7]. They also say doctors need training to use AI properly and that AI should help doctors do their job better, not get in the way[7].\n\nMeanwhile, **China** has announced a big national plan to use AI in healthcare[12]. Starting between November 2025 and March 2026, China will begin planning how to use AI, then test it in 50 hospitals and 500 clinics[12]. This shows that AI in healthcare is becoming important everywhere in the world, not just in the United States.\n\nThe money flowing into healthcare AI companies is huge. The worldwide AI healthcare market is expected to grow from 37.09 billion dollars in 2025 to **701.79 billion dollars by 2034**[8]. However, some people who invest in companies are worried because hospitals are being slow to use AI[5]. Only 3 out of every 10 AI projects that hospitals try actually end up being used for real patients[5]. The problem is that hospitals move slowly because they need to be very careful about patient safety, while AI companies want to move fast[5].\n\nExperts warn that not all AI tools for mental health and wellness have enough proof that they really work[11]. The **American Psychological Association** says that AI chatbots and apps for feeling better need more testing and better rules before people use them[11]. This shows that while AI can help healthcare in many ways, we still need to be careful and test everything thoroughly to keep patients safe."
}
```

New guidance offered for responsible AI use in health care

Solving humanity's healthcare crisis means turning to AI

How Can AI Sentiment Analysis Apply to Complex Medical ...

AI on the Horizon: November 13, 2025

The Patience Gap In Healthcare AI, And What To Do About It

U.S. FDA and CMS Actions on Generative AI-Enabled ...

4 crucial things for Capitol Hill to consider as health AI ...

AI In Healthcare Market to Hit USD 701.79 Billion by 2034

Bonus Features – November 16, 2025 – Half of clinicians ...

Global Health Equity Week Kicks off with a Focus on AI for ...

Artificial intelligence, wellness apps alone cannot solve ...

China Unveils National AI Healthcare Strategy: A Medical ...

Healthcare is changing quickly thanks to **artificial intelligence agents** that can work like doctors[2]. These special AI systems don't just follow instructions - they can think, make choices, and take actions on their own. They can talk to patients, understand their problems, figure out what's wrong with them, and suggest treatments. This is very different from regular AI that just does one specific job.

One of the most exciting examples is **Scope AI**, created by a company called Akido Labs[2]. Scope AI can do what is called an "omni-specialty" medical job, meaning it can help with many different types of medical problems[2]. When a patient talks to Scope AI, it asks questions, learns about their health, and runs that information against millions of real patient cases to find the best diagnosis[2]. Then it suggests tests to run, writes notes about the patient, and even recommends medicines or lifestyle changes[2]. The really important part is that a real human doctor always reviews everything Scope AI suggests before anything is given to the patient[2]. In Akido's clinics in **Southern California**, United States, doctors say they can see four times more patients each day because Scope AI helps them work faster[2]. By the end of 2025, more than 250,000 patients will have used Scope AI for their medical visits[2].

Because AI is becoming such a big part of healthcare, important medical groups are creating rules to make sure it is used safely. The **American Heart Association** released guidance this week that tells hospitals how to pick, test, and watch AI tools to make sure they help patients[1]. The guidance says hospitals should use **four main principles**: making sure AI matches what the hospital wants to do, checking that AI is fair and does not treat some patients worse than others, making sure AI actually helps patients get better, and thinking about costs[1]. The Association found something worrying - only 61 percent of hospitals test their AI tools on their own local patients before using them, and even fewer check to see if the AI treats all patients fairly[1]. Small, rural, and non-academic hospitals have even bigger problems with this[1].

The **FDA**, which is the government group in the United States that approves new medicines and medical devices, is also paying attention to AI[6]. On November 6, 2025, the FDA held a meeting to talk about AI systems that help with mental health[6]. The FDA is trying to figure out the right way to check these AI systems to make sure they are safe and helpful[6]. One of their big worries is that AI keeps changing and learning, so doctors and the FDA need new ways to test it and make sure it stays safe for a long time[6].

The **American Medical Association**, which is a big group of doctors in the United States, just started a new center to help with AI and digital health[7]. They say that doctors must be part of every step of making and using AI in healthcare[7]. The AMA thinks the government needs to create one clear set of rules instead of many different rules, so everyone knows what to do[7]. They also say doctors need training to use AI properly and that AI should help doctors do their job better, not get in the way[7].

Meanwhile, **China** has announced a big national plan to use AI in healthcare[12]. Starting between November 2025 and March 2026, China will begin planning how to use AI, then test it in 50 hospitals and 500 clinics[12]. This shows that AI in healthcare is becoming important everywhere in the world, not just in the United States.

The money flowing into healthcare AI companies is huge. The worldwide AI healthcare market is expected to grow from 37.09 billion dollars in 2025 to **701.79 billion dollars by 2034**[8]. However, some people who invest in companies are worried because hospitals are being slow to use AI[5]. Only 3 out of every 10 AI projects that hospitals try actually end up being used for real patients[5]. The problem is that hospitals move slowly because they need to be very careful about patient safety, while AI companies want to move fast[5].

Experts warn that not all AI tools for mental health and wellness have enough proof that they really work[11]. The **American Psychological Association** says that AI chatbots and apps for feeling better need more testing and better rules before people use them[11]. This shows that while AI can help healthcare in many ways, we still need to be careful and test everything thoroughly to keep patients safe.

Healthcare systems around the world are increasingly using **artificial intelligence agents** that can think and make decisions like doctors[2]. One major example is Scope AI, an AI system created by Akido Labs that can talk to patients, ask questions, figure out what disease they might have, and create treatment plans[2]. A human doctor still checks everything before it gets used, which makes it safe[2]. By the end of 2025, over 250,000 patients will use Scope AI visits[2].

Doctors and leaders are working hard to make sure AI is used the right way. The **American Heart Association** released new rules to help hospitals use AI safely and fairly[1]. They say AI tools need to be tested carefully before hospitals use them, especially to make sure they work well for all different kinds of patients[1]. The **FDA** (the government group that approves medicines and medical devices in the United States) is also making new rules for AI systems that help with mental health[6].

Companies that sell AI are getting lots of money to create these tools. The AI healthcare market is growing very fast and might reach **701.79 billion dollars by 2034**, up from 37.09 billion dollars in 2025[8]. However, some experts warn that not all the AI tools being created will actually work well in real hospitals[5]. **China** recently announced a big plan to use AI in healthcare, with testing starting soon in 50 hospitals[12]. Even though AI is exciting and can help many patients, doctors say that people must still be part of the decision-making process to keep patients safe[7].

Healthcare Weekly AI News

Specific Topics