February 2017 – Scientists Talk Funny

Computer algorithm predicts outcome for leukemia patients

Researchers have developed a machine-learning computer algorithm that predicts the health outcome of patients with acute myeloid leukemia — identifying who is likely to relapse or go into remission after treatment.

Acute myeloid leukemia (AML) is a cancer characterized by the rapid growth of abnormal white blood cells that build up in the bone marrow and interfere with the production of normal blood cells.

A standard tool used for AML diagnosis and treatment monitoring is flow cytometry, which measures the physical and chemical characteristics of cells in a blood or bone marrow sample to identify malignant leukemic cells. The tool can even detect residual levels of the disease after treatment.

Unfortunately, scientists typically analyze this flow cytometry data using a time-consuming manual process. Now, researchers from Purdue University and Roswell Park Cancer Institute believe they have developed a machine-learning computer algorithm that can extract information from the data better than humans.

“Machine learning is not about modeling data. It’s about extracting knowledge from the data you have so you can build a powerful, intuitive tool that can make predictions about future data that the computer has not previously seen — the machine is learning, not memorizing — and that’s what we did,” said Murat Dundar, PhD, associate processor at Indiana University-Purdue University, in a recent news release.

The research team trained their computer algorithm using bone marrow data and medical histories of AML patients along with blood data from healthy individuals. They then tested the algorithm using data collected from 36 additional AML patients.

In addition to being able to differentiate between normal and abnormal samples, they were able to use the flow cytometry bone marrow data to predict patient outcome — with between 90 and 100 percent accuracy — as recently reported in IEEE Transactions on Biomedical Engineering.

Although more work is needed, the researchers hope their algorithm will improve monitoring of treatment response and enable early detection of disease progression.

Dudar explained in the release:

“It’s pretty straightforward to teach a computer to recognize AML. … What was challenging was to go beyond that work and teach the computer to accurately predict the direction of change in disease progression in AML patients, interpreting new data to predict the unknown: which new AML patients will go into remission and which will relapse.”

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

New blood test could detect early-stage pancreatic cancer

Pancreatic cancer is one of the leading causes of cancer death, because it is seldom detected before the disease has spread to other organs. Only 8 percent of people with pancreatic cancer survive five or more years after diagnosis.

Now, researchers hope to change this bleak scenario with an improved blood test that can detect early-stage pancreatic cancer. A multi-institutional team led by Tony Hu, PhD, an associate professor at Arizona State University, recently reported on their results in Nature Biomedical Engineering.

The researchers first identified the presence of a protein in the blood, called ephrin type-A receptor (EphA2), which is overexpressed by pancreatic tumors. Next, they developed a biosensor using gold nanoparticles that selectively bind to EphA2, changing their light emitting properties. This allowed the team to quantify the amount of EphA2 in a blood sample to see if it is overexpressed.

They validated their biosensor in a pilot study involving 48 healthy people, 59 patients with stage I-III pancreatic cancer and 48 patients with chronic pancreas inflammation. The later condition is often confused with pancreatic cancer using existing diagnostic tests like ultrasound.

The biosensor was able to accurately identify the patients with pancreatic cancer — even those with early stage disease — as well as the patients with chronic pancreas inflammation. If these results are validated with a larger clinical trial, the blood test could screen for pancreatic cancer and could be adapted for other diseases.

“We are now working on lung cancer and lymphoma and have very positive results,” Hu said in a recent news story. “In addition to cancer, we are conducting a project on tuberculosis diagnosis. Theoretically this test could be applied to any type of disease.”

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

Unable to smell? One Stanford researcher is working to improve therapies

adult-19033_1920 — Photo by PublicDomainPictures

I don’t often think about my sense of smell, unless I’m given a fragrant flower or walk past someone smoking. But the ability to smell is both critical and underappreciated, according to Zara Patel, MD, a Stanford assistant professor of otolaryngology, head and neck surgery.

A smell begins when a molecule — say, from a flower — stimulates the olfactory nerve cells found high up in the nose. These nerve cells then send information to the brain, where the specific smell is identified. Anything that interferes with these processes, such as nasal congestion or damage to the nerve cells, can lead to a loss of smell.

I recently spoke with Patel about the loss of the sense of smell, a condition known as anosmia.

How does losing the sense of smell impact patients?

“If asked which sense they’d give up first, most people would likely choose their sense of smell. It’s only after the loss of olfaction that its significant impact on our lives is appreciated. Our sense of smell plays a key role in a vast array of basic human interactions, such as what attracts us to sexual partners, what keeps us in committed relationships and how maternal bonding occurs with newborns. It’s also one of our most basic protective mechanisms that allows us to wake up in the midst of a fire and prevents us from eating spoiled food. And importantly — keeping in mind that our ability to taste is highly dependent on our ability to smell — the inability to enjoy food and related social activities often causes social isolation, depression and malnutrition.”

What causes olfactory loss?

“There are over 100 reasons why people can lose their sense of smell. However, the majority of people lose it from sinonasal inflammatory disease, post-viral infections, traumas or tumors. Unfortunately, olfactory loss is often of “idiopathic” origin, meaning we just don’t know what caused it. That is why research in this area is so important.

It’s also important to be treated as early as possible. It is always frustrating to see someone who lost their sense of smell over a year ago, but they weren’t referred to me at the time or were told that nothing could be done. Those are missed opportunities that will negatively impact those patients for the rest of their lives.”

How do you treat patients who can no longer smell?

“The treatment really depends on the reason for loss, and may include surgery or medications. For those who lose the ability to smell after trauma, post-viral infection or when we don’t know why it happened, olfactory training can be used, which is a very simple protocol that patients can do at home. The patients smell several essentials oils in a structured way twice a day, every day, over a long period of time. The oils — rose, eucalyptus, clove and lemon —stimulate different types of olfactory receptor cells in the nose. Although it does not help everyone, it has been shown to be effective in 30 to 50 percent of patients, across multiple origins of loss.

We don’t have an exact understanding of how and why it works. However, a study using functional MRI observed a change in how the brain responds to odors before and after olfactory training. Before the training, there was a chaotic array of random areas lighting up in the brain. After the training, the images showed a renewed pathway to the olfaction center in the brain. We also know that the olfactory nerve has an inherent ability to regenerate. We’re trying to take advantage of this fact and ‘switch on’ those regenerative cells.

I have many patients who have benefited from olfactory training, including some who need their sense of smell for their livelihood — such as chefs or wilderness guides. Being able to get that sense back has allowed them to continue doing what they’re passionate about and has increased their quality of life.”

What are you working on now?

“Although olfactory training has allowed us to help more patients, 30 to 50 percent improvement is still quite low and certainly not the final answer. That’s why the research I’m currently doing has me excited about the potential of using both stem cells and neurostimulation to advance this field. I hope to soon be able to offer alternative interventions to these patients.”

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

Stanford researchers develop simulations to improve heart surgeries

MRI or CT scans provide physicians with a detailed picture of their patients’ internal anatomy. Heart surgeons often use these images to plan surgeries.

Unfortunately, these anatomical images don’t show how the blood is flowing through the vessels — which is critical, according to Alison Marsden, PhD, a Stanford associate professor of pediatrics and of bioengineering. In the video above, she explains that many surgeons currently use a pencil and paper to sketch out their surgical plan based on the patient’s images. She hopes to change this.

Marsden and her colleagues at Stanford’s Cardiovascular Biomechanics Computational Lab are developing a new technique — using imaging data and specialized simulation software — to predict what is likely to happen during heart surgery.

“What we’re trying to do is bring in that missing piece of what are these detailed blood flow patterns and what might happen if we go in and make an intervention, for example, opening up a blocked blood vessel or putting in a bypass graft,” Marsden said in a recent Stanford Engineering news story.

Their open source software, called SimVascular, loads the imaging data, constructs a 3D anatomical model of the heart and then simulates the patient’s blood flow. It has already been used to help design the surgical plan for several babies born with a severe form of congenital heart disease, Marsden said. However, more research is needed to determine whether the technique improves patient outcomes before it can be widely used in the clinic.

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

What you need to know about e-cigarettes

E-cigarettes are extremely popular with millions of middle and high school students across the United States. Kids love the flavors — like strawberry, bubble gum, chocolate cake and cotton candy — and blowing vapor into rings. And, they are inundated with ads that tout e-cigarettes as cool, harmless alternatives to cigarettes.

But, not surprisingly, e-cigarettes aren’t really safe. A recent University of California news story outlines ten important facts about e-cigarettes, including how they can harm your health.

One of the biggest health concerns is that e-cigarettes contain nicotine, which is addictive and can lead to the use of traditional cigarettes. “A lot of kids who take up [nicotine-free] vaping are at low risk for smoking, but once they start using e-cigarettes, they are three to four times more likely to start using cigarettes,” said Stanton Glantz, PhD, a tobacco researcher at the University of California, San Francisco, in the article.

In addition, e-cigarettes can contain other harmful ingredients, including:

Ultrafine particles that can trigger inflammatory problems and lead to heart and lung disease
Toxic flavorings that are linked to serious lung disease
Volatile organic compounds
Heavy metals, such as nickel, tin and lead

Stanford’s Bonnie Halpern-Felsher, PhD, a developmental psychologist who has studied tobacco use, also commented in the piece:

“Youth are definitely using e-cigarettes because they think they are cool… Adolescents and young adults don’t know a lot about e-cigarettes. They think it’s just water or water vapor. They don’t understand it’s an aerosol. They don’t understand that e-cigarettes can have nicotine. They don’t understand that flavorants themselves can be harmful.”

Furthermore, when e-cigarette users exhale the mainstream vapor containing these toxins, they can cause secondhand health effects.

The article discusses other hazards as well, including the possibility of battery explosion, and the products’ mixed record on helping smokers quit. It concluded with a call for more research to better understand the long-term health effects of e-cigarettes.

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

Stanford researchers map brain circuitry affected by Parkinson’s disease

In the brain, neurons never work alone. Instead, critical functions of the nervous system are orchestrated by interconnected networks of neurons distributed across the brain — such as the circuit responsible for motor control.

Researchers are trying to map out these neural circuits to understand how disease or injury disrupts healthy brain cell communication. For instance, neuroscientists are investigating how Parkinson’s disease causes malfunctions in the neural pathways that control motion.

Now, Stanford researchers have developed a new brain mapping technique that reveals the circuitry associated with Parkinson’s tremors, a hallmark of the disease. The multi-disciplinary team turned on specific types of neurons and observed how this affected the entire brain, which allowed them to map out the associated neural circuit.

Specifically, they performed rat studies using optogenetics to modify and turn on specific types of neurons in response to light and functional MRI to measure the resulting brain activity based on changes in blood flow. These data were then computationally modeled to map out the neural circuit and determine its function.

The research was led by Jin Hyang Lee, PhD, a Stanford electrical engineer who is an assistant professor of neurology and neurological sciences, of neurosurgery and of bioengineering. A recent Stanford News release explains the results:

“Testing her approach on rats, Lee probed two different types of neurons known to be involved in Parkinson’s disease — although it wasn’t known exactly how. Her team found that one type of neuron activated a pathway that called for greater motion while the other activated a signal for less motion. Lee’s team then designed a computational approach to draw circuit diagrams that underlie these neuron-specific brain circuit functions.”

“This is the first time anyone has shown how different neuron types form distinct whole brain circuits with opposite outcomes,” Lee said in the release.

Lee hopes their research will help improve treatments for Parkinson’s disease by providing a more precise understanding of how neurons work to control motion. In the long run, she also thinks their new brain mapping technique can be used to help design better therapies for other brain diseases.

This is a reposting of my Scope blog story, courtesy of Stanford School of Medicine.

Share this:

Like this:

Share this:

Like this:

Share this:

Like this:

Share this:

Like this:

Share this:

Like this:

Share this:

Like this: