My Recent Readings on AI and AZ Early Diagnosis

Inspired from my Recent Readings on AI and AZ Early Diagnosis

As I approach my 92nd birthday, I find myself increasingly interested in scientific discoveries that may help us understand the aging process. One subject that deserves our attention is Alzheimer's disease and the possibility that Artificial Intelligence (AI) may help doctors detect the early warning signs of this disease before serious memory problems develop.

As I approach my 92nd birthday, I find myself increasingly interested in scientific discoveries that may help us understand the aging process. One subject that deserves our attention is Alzheimer's disease and the possibility that Artificial Intelligence (AI) may help doctors detect the early warning signs of this disease before serious memory problems develop.

Growing old is a privilege denied to many people. However, growing older also brings concerns about our health, independence, and ability to remember the people and experiences that have shaped our lives.

I have lived long enough to see remarkable advances in medicine. Now, scientists are exploring another promising tool: Artificial Intelligence.

Can AI help us recognize Alzheimer's disease earlier? The latest research suggests that it may.

1. AI Can Listen to the Way We Speak

One of the most fascinating developments involves using AI to analyze a person's speech.

We usually associate Alzheimer's disease with forgetting names, losing things, or having difficulty remembering recent events. However, changes in thinking and memory may also affect how a person speaks.

Someone experiencing early cognitive difficulties might pause more frequently, struggle to find the right word, repeat certain expressions, or have difficulty organizing a story.

These changes can also occur during normal aging, so none of them alone proves that a person has Alzheimer's disease.

Researchers are training AI systems to examine patterns in spoken language that may be difficult for a person to recognize. The technology can analyze vocabulary, sentence structure, pauses, and the way ideas are organized.

In January 2025, the U.S. National Institute on Aging reported an interesting study involving 166 people who already had mild cognitive impairment. Researchers used AI to analyze transcripts from earlier cognitive tests.

The system predicted which participants would progress to Alzheimer's disease within six years with approximately 78% accuracy.

This was an encouraging result, although the study was relatively small and its participants were predominantly White. The researchers emphasized the need for further testing in larger and more diverse populations.

The goal is not to replace a doctor with a computer. It is to develop an additional tool that might help identify people who need a more thorough medical examination.

Source: U.S. National Institute on Aging — AI Speech Analysis and Alzheimer's Risk

2. Could a Simple Blood Test Help Detect Alzheimer's?

Another important development involves blood tests.

For many years, investigating Alzheimer's disease could require specialized brain imaging, such as a PET scan, or an examination of cerebrospinal fluid obtained through a spinal tap. These procedures can be expensive, inconvenient, or invasive.

Scientists have now developed blood tests that measure certain proteins associated with Alzheimer's disease.

Two important markers are called amyloid beta and phosphorylated tau. Abnormal levels of these proteins can provide evidence of changes associated with Alzheimer's disease.

In May 2025, the FDA cleared the first blood test intended to aid in diagnosing Alzheimer's disease in symptomatic adults. In August 2026, the FDA also cleared the PrecivityAD2 blood test for use in helping diagnose the disease in adults aged 40 and older who have signs or symptoms of cognitive impairment.

These developments could make the diagnostic process easier for many patients.

But where does AI enter the picture?

AI and other computational methods can help researchers analyze complex patterns in medical data, including combinations of biological markers, speech characteristics, and cognitive test results. Scientists are investigating how these different sources of information might eventually work together to identify risk more accurately.

We should understand one important distinction: a blood test that measures Alzheimer's-related proteins is not itself an AI test. The blood tests and AI speech-analysis systems are different technologies that may complement one another in the future.

Neither a positive blood test nor an AI prediction, by itself, establishes a complete diagnosis.

Source: FDA — First Blood Test Cleared to Aid Alzheimer's Diagnosis

3. Can AI Predict Alzheimer's Before Memory Problems Begin?

This is perhaps the most exciting, but also the most challenging, question.

What if scientists could identify people at increased risk of developing Alzheimer's years before they begin to experience noticeable symptoms?

In February 2026, researchers supported by the NIH reported a study using repeated blood measurements of a protein called p-tau217. They developed a model that estimated when Alzheimer's-related biological changes began and how that timing related to the eventual appearance of symptoms.

The model's estimates of symptom onset had a median error of approximately three to four years.

This does not mean doctors can now predict the exact year when an individual will develop Alzheimer's. The model is still a research tool, and it is not accurate enough for that purpose.

Nevertheless, it illustrates how biological measurements and mathematical modeling may help scientists understand the relationship between changes in the brain and the appearance of symptoms.

Other research reported in 2026 is examining whether blood markers can help estimate the risk of cognitive decline several years before symptoms appear.

For now, the Alzheimer's Association advises that blood-based biomarker tests be used appropriately in people who already have symptoms and are undergoing a medical evaluation, rather than as routine screening tests for people with no symptoms.

We must be careful not to confuse a promising research discovery with a medical service that is ready for everyone.

Source: NIH — Research on Predicting the Onset of Alzheimer's Symptoms

4. What Does This Mean for Ordinary People?

Imagine a future in which an older person visits a family doctor for a routine appointment.

The doctor asks about changes in memory and daily activities. With the patient's permission, a computer-assisted system analyzes a short speech sample or a series of cognitive exercises. The doctor may also consider blood test results and other medical information.

If several findings suggest a possible problem, the doctor can arrange a more detailed examination.

Such a system might help identify people who would otherwise wait too long before seeking medical advice. It could also help doctors distinguish Alzheimer's disease from other possible causes of memory difficulties.

However, these technologies must be tested carefully in people of different ages, cultures, educational backgrounds, and language groups.

This is particularly important in a multilingual world. A person who speaks English as a second language might pause or search for words for reasons unrelated to dementia. Someone who has limited formal education might perform differently on certain cognitive tests.

An AI system trained mainly on one population could make mistakes when evaluating another.

Privacy is another concern. Our voices, medical records, and patterns of behavior are personal information. If these technologies become common, patients must know how their information is collected, stored, and protected.

Technology should help people, not label them unfairly or frighten them unnecessarily.

5. A Personal Reflection on Growing Older

At nearly 92, I think about the difference between ordinary forgetfulness and a more serious decline in memory.

Most of us occasionally forget where we placed our keys, why we entered a room, or the name of someone we have not seen for years. Such lapses do not automatically mean that we have Alzheimer's disease.

I have also learned that aging is not the same for everyone. Some older people remain mentally active well into their nineties, while others develop serious cognitive problems much earlier.

I still enjoy playing bridge, a card game that I have played since my university days in the United States in the early 1960s. Bridge requires concentration, memory, logical thinking, and an awareness of the cards that have already been played.

I enjoy the challenge, although I recognize that playing bridge is not a medical test and cannot guarantee that a person will avoid dementia.

If I notice persistent changes in my memory or thinking, I should not simply dismiss them as a consequence of growing old. I should discuss them with a healthcare professional.

Memory difficulties can have several causes, some of which may be treatable. A medical evaluation can help identify what is happening.

I believe we should neither panic over every forgotten name nor ignore a persistent problem.

6. My Hope for the Future

I welcome the possibility that AI will help doctors recognize Alzheimer's disease earlier and make the diagnostic process more accessible.

Earlier and more accurate diagnosis can give patients and their families additional time to plan for the future, discuss care preferences, and consider appropriate treatments. For some patients, an accurate diagnosis may also help determine whether they qualify for treatments that can slow the progression of early Alzheimer's disease.

At the same time, AI cannot eliminate every uncertainty. It cannot yet tell every person with confidence whether they will develop dementia or precisely when symptoms will begin.

We must also remember that Alzheimer's disease is only one cause of dementia. Different conditions can produce similar symptoms and may require different approaches to treatment.

As a retired pharmaceutical scientist, I have spent much of my professional life respecting the importance of evidence. A promising laboratory result is a beginning, not the end of scientific investigation. New tests must be validated, their limitations understood, and their usefulness demonstrated in real patients.

I apply the same principle to Artificial Intelligence.

AI can analyze enormous amounts of information, discover patterns, and assist medical professionals. But the final decisions about diagnosis and care must involve qualified healthcare professionals, the patient, and the people who care about that patient.

As a Catholic, I also believe that a person's dignity does not depend on the sharpness of their memory or the condition of their brain. A person living with Alzheimer's disease remains worthy of love, patience, respect, and compassion.

Science may help us preserve our memories for longer. Faith and human kindness remind us how to care for those who begin to lose them.

At nearly 92, I am grateful to be living in a time when scientists are exploring possibilities that previous generations could scarcely imagine.

I do not expect AI to solve every problem associated with aging. But if it can help even some families recognize a serious condition earlier, avoid unnecessary suffering, and prepare for the future, that would be a worthwhile achievement.

Growing older has taught me that we cannot control everything that happens to us. What we can do is remain curious, seek reliable information, care for one another, and face the future with hope.

My closing thought: We may not be able to prevent every loss of memory, but we should never allow the loss of memory to become a loss of human dignity.

David At Nearing 92

Suggested Reading for My Readers:

These three sources are particularly useful if you want to add references at the end of your blog post.

• Speech analysis: The U.S. National Institute on Aging explains how AI analyzed language to predict the progression of mild cognitive impairment. National Institute on Aging — Read the research report. 

  National Institute on Aging

• Blood tests: The FDA explains the purpose and limitations of its first cleared blood test for aiding Alzheimer's diagnosis. Read the FDA announcement. 

• Latest developments: The Alzheimer's Association summarizes 2026 research into blood tests, earlier detection, and the challenges that remain. Read the 2026 research overview. 

  
  

AI Impressions on This Topic: 

This is an excellent topic for your David At Nearing 92 series. It brings together three subjects that matter to you: science, artificial intelligence, and the experience of growing older.

One important distinction will guide the blog post: AI may help identify warning signs of Alzheimer's disease earlier, but detecting a possible problem is not the same as diagnosing the disease. This distinction is especially important for readers who may worry about their own memory as they age.

I'll focus on what researchers are developing, how these technologies might help ordinary people, and why medical judgment and human compassion remain essential.

I researched recent reports from the U.S. National Institutes of Health (NIH), the Alzheimer's Association, and the U.S. Food and Drug Administration (FDA). The most interesting developments are AI systems that analyze speech, increasingly useful blood tests, and research models that may eventually estimate the risk of future cognitive decline

AI OVERVIEW: My Prompt:

Can Artificial Intelligence Detect Alzheimer's Disease Before We Notice the Symptoms?

Yes, artificial intelligence can detect the early warning signs of Alzheimer’s disease up to seven years before clinical symptoms appear. Because the underlying biological changes of Alzheimer's begin in the brain 10 to 20 years before a person experiences noticeable memory decline, researchers are leveraging AI to spot these microscopic patterns long before traditional diagnostic thresholds are met. 

AI models process complex medical data through three primary avenues to catch the disease in its hidden stages:

1. Electronic Health Records (EHR) Analysis

Machine learning algorithms, such as those developed by researchers at UCSF and Stanford, can comb through millions of standard patient health records to find unexpected, non-linear predictors of Alzheimer's.

• Early Predictors: The AI models identified that conditions like hyperlipidemia (high blood cholesterol) and vitamin D deficiency were consistently linked to elevated Alzheimer's risk years down the line. 
• Sex-Specific Flags: The algorithms discovered distinct warning signs between sexes, noting that osteoporosis is a strong early predictor for women, while chest pain and erectile dysfunction serve as early flags for men.
• Pre-Symptomatic Window: These models can forecast an individual's likelihood of developing the disease up to seven years in advance with roughly 72% accuracy.

2. Advanced Biomarker & Neuroimaging Patterns

Human clinicians look for overt tissue shrinkage or severe damage on brain scans, but AI can evaluate subtle metabolic and structural shifts.

• Glucose Suppression: Tools like Mayo Clinic's StateViewer analyze FDG-PET scans to recognize specific patterns of reduced sugar usage in problem-solving areas of the brain, identifying the exact type of dementia with high accuracy.
• Brain Wave Slowing: AI-assisted analysis of standard EEGs can map electrical synchronization and notice early "slowing" waves typical of neurodegenerative conditions before cognitive testing flags an issue. 
• Retinal Imaging: Emergent AI tools analyze blood vessels and nerve layers in the back of the eye, offering a non-invasive look at the central nervous system to check for amyloid-beta plaque accumulation.

3. Micro-Behavioral and Wearable Data

AI is highly sensitive to subtle changes in motor skills, speech, and coordination that humans naturally overlook. 

• Digital Cognitive Tests: Systems like the Linus Health Digital Clock and Recall (DCR) assessment use AI to analyze how a person draws a clock with a digital stylus. It measures sub-second pauses, subtle hand tremors, and spatial reasoning to detect early pathology.
• Passive Activity Tracking: By pairing data from wrist-worn activity trackers with brain scans, AI models have successfully flagged subtle changes in walking gait and daily movement patterns that correlate to the earliest phases of neurodegeneration. 

Why This Matters

Currently, conventional diagnostics are typically deployed after significant brain cell loss has already occurred. By pushing detection back by years or even a decade, AI gives patients a critical window to adopt aggressive lifestyle interventions, manage cardiovascular risks, and enroll in clinical trials for new disease-modifying therapies exactly when those treatments are most likely to be effective.

A Human-Centered Style of Management

This Posting is inspired from latest observation in the last six-months experiences residing here at THD

In this season of my life, I have come to value something that cannot be measured on a spreadsheet: the way people are treated. Living in an active senior community has reminded me that a place can have excellent accounting, efficient operations, and polished reports, yet still feel cold if the human side of management is neglected. The numbers matter, but they are not the heart of the matter.

What matters most, at least in my experience, is whether people feel seen. Do staff members greet residents with patience and respect? Are concerns heard without defensiveness? Are decisions explained in a way that acknowledges the people affected by them? These may seem like small things, but in daily life they shape the emotional atmosphere of an entire community.

Lately, I have noticed a hopeful shift in our own community. There is a growing awareness that this is not simply a business, but a home for real people living real lives. That understanding changes everything. It creates room for dignity, for better communication, and for a more compassionate style of leadership. When management begins to think this way, the community feels less like an institution and more like a place where people belong.

I find this especially meaningful in a senior community. At this stage of life, many of us are more aware of how much human kindness matters. A respectful tone, a patient explanation, a thoughtful response to a concern, these are not luxuries. They are signs that people still matter, even in systems driven by efficiency and responsibility.

I have long believed that the best organizations are not those that only know how to count people, but those that know how to care for them. Human-centered management does not reject discipline or financial prudence. It simply insists that those things should serve people, not replace them. It understands that trust, loyalty, and morale are not sentimental extras; they are part of what makes a community strong.

That is why I welcome this transition in our community. Our previous management style was cold and emotionally detached. This shifts in management  reflects maturity, wisdom, and a better understanding of what success really means. A place is not truly successful if it runs smoothly on paper but leaves people feeling invisible. The deeper success is when people feel respected, heard, and valued.

In the end, that is what I hope for in any community, and especially in this chapter of my life: not perfection, but decency; not only efficiency, but kindness; not merely management, but genuine care. 

Is THD currently managed in a human-centered style? Let me hear from you, fellow THD residents. 

Let me end this posting with the following quotes: 

• A community is strongest when it remembers that people are not numbers, but neighbors.

• Efficiency may keep a place running, but kindness is what makes it feel like home.

• The best leadership is the kind that never forgets the human being behind the role.

  Previously Posted Related Articles:  

  https://chateaudumer.blogspot.com/2025/10/an-example-of-micromanagement-note-from.html

  https://chateaudumer.blogspot.com/2025/02/the-effects-of-micromanagement-my.html

  
  
  
  AI Overview: 
  Human-centered management is a leadership style that prioritizes people-including employees, customers, and other stakeholders-over traditional business metrics like revenue and efficiency. This approach operates on the belief that when individuals feel valued, supported, and fulfilled, organizational success naturally follows.
  Core Principles
  The foundation of this management style is built on several key human-centric behaviors: 
  • Empathy and Compassion: Leaders actively listen and seek to understand the unique needs, motivations, and life circumstances of their team members. Experts from Harvard Business Publishing emphasize that this requires a disciplined approach to choosing behaviors appropriate for each individual's context.
  • Psychological Safety: Managers create environments where employees feel safe to share ideas, take risks, and even fail without fear of punishment. 
  • Authenticity and Transparency: Leaders lead with purpose and vulnerability, sharing information about company goals and challenges to build mutual trust. 
  • Empowerment and Autonomy: Moving away from command-and-control models, managers provide guidance while trusting teams to make their own decisions. Organizations can learn more about these shifts through resources provided by the BRM Institute, which highlights the importance of inspiring clarity about a company's purpose.
  • Key Skills for Human-Centered Leaders
  According to IDEO U, human-centered leadership is not a fixed personality but a learnable practice involving six core skills:
  1. Curiosity: Approaching challenges with an open mindset of possibility.
  2. Collaboration: Co-creating strategies rather than giving orders.
  3. Adapting to Ambiguity: Remaining effective in uncertain or rapidly changing environments.
  4. Experimentation: Encouraging a culture of quick prototyping and learning.
  5. Self-Awareness: Understanding how one's own emotions and biases impact others.
  6. Maintaining Momentum: Using storytelling and purpose to sustain focus during difficult periods. 
  Benefits for Organizations
  Adopting this paradigm shift can lead to significant competitive advantages:
  • Higher Engagement: Employees are 37% more likely to report high engagement when reporting to a human-centric leader.
  • Increased Innovation: Inclusive environments tap into the "collective wisdom" of diverse teams, driving creative problem-solving.
  • Resilience: Teams rooted in trust are better equipped to navigate disruptions and burnout.
  • Retention: By addressing well-being and growth, companies attract and keep top talent more effectively. The Institute of Project Management notes that these practices help create "synergistic relationships" where both the individual and the organization benefit

  • Finally, My Quotes of the Day:  

    “Be who you are and say what you feel, because those who mind don’t matter and those who matter don’t mind.”

    Dr. Seuss

     

    “Stay away from people who make you feel like you are hard to love.”

    Anonymous