Improvement on the effect of anti-cancer drugs

Share on:

Medical technology developed at the Technion improves the effect of anti-cancer drugs:

The technology, based on the Nano-Ghosts platform, makes it possible to reduce the drug dose a million-fold without reducing its efficacy

Prof. Marcelle Machluf

Researchers in the Faculty of Biotechnology and Food Engineering at the Technion developed a technology that inhibits development of melanoma using one-millionth of the active ingredient. The study, published in Advanced Functional Materials, was led by Prof. Marcelle Machluf, dean of the faculty, and PhD student Lior Levy.

Immunotherapy in action. This development is a significant breakthrough in the field of immunotherapy – an innovative medical approach that has become one of the most promising trends in cancer treatment. The approach is based on the ability of the body’s own immune system to destroy cancer cells. This system can do that more accurately and specifically than synthetic anti-cancer drugs. However, since the malignant tumor is heterogeneous and evasive, it can sometimes fool the immune system, and this is where science enters the picture, with new tools that help the immune system deal with this challenge.

Lior Levy

TRAIL protein. At the core of this new development is a protein called TRAIL, which exists in the body’s immune system and knows how to induce apoptosis (programmed cell death) of cancer cells. In other words, it is a Tumor Necrosis Factor (TNF). Another advantage: it is selective, meaning it only affects cancer cells, a highly desirable feature in anti-cancer treatment. The application of TRAIL in immunotherapy has so far encountered various technical challenges, including the absorption of the protein in the body, its distribution (pharmacokinetics), and the fact that it does not survive for very long. This study offers a solution to these problems.

Nano-Ghosts technology. The development presented in the Technion researchers’ article is based on original technology developed by Prof. Machluf in her years at the Technion: Nano-Ghosts. The platform is produced by emptying specific biological cells (mesenchymal stem cells) in a way that leaves only the cell membrane and reducing their size to a nanometer scale. Any drug can be inserted into the membrane and injected directly into the bloodstream. Because the body’s immune system treats nano-ghosts as natural cells, it delivers them to the affected site. They do not release the drug on the way, and therefore do not harm healthy tissue. They target the malignant tissue, where they deliver the drug into the tumor cells.

Integration. The study integrates the three aforementioned factors: the immunotherapy concept, the TRAIL protein, and the Nano-Ghost technology developed by Prof. Machluf. The result: a drug delivery system with the active protein on its outer layer, which allows reduction of the drug dosage by a factor of a million while maintaining the same treatment effect.

A schematic description of preparing Nano-Ghosts from cells that underwent genetic or metabolic manipulation and now carry the TRAIL protein. These cells, with Nano-Ghost targeting for cancer and with the TRAIL protein, can reach the cancer site and fight effectively while using one millionth the concentration of the active ingredient required without this system.

According to Prof. Machluf, “this integration turns the Nano-Ghost platform from a “taxi” that delivers the drug to the target into a “tank” that participates in the war. The integrated platform delivers the drug to the tumor and enables a significant reduction in drug dosage yet still does the job. We also showed that our method does not harm healthy cells.”

The technology was demonstrated on cells in the lab and on human cancer cells in mice. The researchers estimate that this new strategy, which was demonstrated in their study on a melanoma model, will also be effective on other types of cancer.


For the article in Advanced Functional Materials click here

AI improves the accuracy of antibiotic selection for the treatment of UTI

Share on:

Maccabi Healthcare Services and the Technion: AI improves the accuracy of antibiotic selection for the treatment of UTI

Shortly after the AI-based advanced tool was introduced, there was a 35% drop in the need to switch antibiotics. The technology was developed by the Technion and Maccabi Healthcare Services

Doctors at Maccabi Healthcare Services have recently begun to work with an AI-based predictive algorithm developed by the Technion – Israel Institute of Technology together with KSM (Kahn-Sagol-Maccabi), the Maccabi Research and Innovation Center. The new algorithm advises doctors in the process of deciding on personalized antibiotic treatment for patients.

The first diagnosis on which Maccabi chose to focus is urinary tract infection (UTI) – the most common bacterial infection among women. Around 30% of the females suffer from the infection at least once during their lifetime, and up to 10% experience recurrent infections. Until now, in most cases general treatment has been administered based on clinical guidelines and medical judgment. Sometimes, the bacteria prove to be antibiotic resistant, resulting in the need to change the treatment plan.

Since the new algorithm was introduced, Maccabi doctors have treated tens of thousands of cases, and there has been a drop of around 35% in the need to switch antibiotics following the development of bacterial resistance to the drug prescribed. This is significant because accuracy in the choice of antibiotic is far greater thanks to the new technology. In light of the success of this new development in the treatment of UTI, Maccabi has begun working on the development of additional detection systems that will help to contend with other infectious diseases that require personalized treatment with antibiotics.

How does it work?

The automated system recommends the most suitable antibiotic treatment for the patient to the doctor, based on clinical guidelines and other criteria such as age, gender, pregnancy status, residence in an assisted living facility, and personal history of UTI and antibiotics administered.

The unique algorithm was developed by Professor Roy Kishony and Dr. Idan Yelin of the Technion Faculty of Biology, in cooperation with KSM, Maccabi’s Research and Innovation Center, headed by Dr. Tal Patalon, and was introduced and implemented among Maccabi’s doctors by the HMO’s Medical Informatics team and Chief Physician’s Department. According to Prof. Kishony, “The algorithm we developed together with Maccabi’s experts is a major milestone in personalized medicine on the way to AI-based antibiotic treatments, which are personally tailored to the patient according to the prediction of treatment response and mitigate the development of resistant bacteria.”

Dr. Shira Greenfield, Director of Medical Informatics at Maccabi Healthcare Services, said, “The significance of administering personalized antibiotic treatment is that it lowers the risk of antibiotic resistance developing – a global problem which all healthcare entities are working to solve.”

We’re Losing Oxygen, and It’s Great!

Share on:

We’re Losing Oxygen, and It’s Great!

 Researchers in the Technion Department of Materials Science and Engineering have succeeded in changing a material’s electrical properties by vacating an oxygen atom from the original structure. Possible applications include electronic-device miniaturization and radiation detection

What do ultrasound imaging of a fetus, cellular mobile communication, micro motors, and low-energy-consumption computer memories have in common? All of these technologies are based on ferroelectric materials, which are characterized by a strong correlation between their atomic structure and the electrical and mechanical properties.


Dr. Hemaprabha Elangovan, Assistant Professor Yachin Ivry and Ph.D. student Maya Barzilay

Technion – Israel Institute of Technology researchers have succeeded in changing the properties of ferroelectric materials by vacating a single oxygen atom from the original structure. The breakthrough could pave the way for the development of new technologies. The research was headed by Assistant Professor Yachin Ivry of the Department of Materials Science and Engineering, accompanied by postdoctoral researcher Dr. Hemaprabha Elangovan and Ph.D. student Maya Barzilay, and was published in ACS Nano. It is noted that engineering an individual oxygen vacancy poses a considerable challenge due to the light weight of oxygen atoms.

In ferroelectric materials, a slight shift of the atoms causes significant changes in the electric field and in the contraction or expansion of the material. This effect is the result of the fact that the basic repeating unit in the material contains atoms that are organized in an asymmetric structure.

In order to explain this further, the researchers use the seminal ferroelectric material, barium titanate, the atoms of which form a cubic-like lattice structure. In these materials, a unique phenomenon occurs: the titanium atom draws away from the oxygen atoms. Since titanium is positively charged and oxygen is negatively charged, this separation creates polarization, or in other words, an electric dipole moment.

In the micrograph: Image of the structure before (on the right) and after (left) removing an oxygen atom.

A cubic lattice has six faces, so the charged atoms move into one of six possibilities. In different parts of the material, a large number of neighboring atoms shift in the same direction, and polarization in each such area, which is known as a ferroelectric domain, is uniform.  Traditional technologies are based on the electric field created in those domains. However, in recent years, a great deal of effort has been directed at minimizing the device size and using the borders, or walls, between the domains rather than the domains themselves, and thus converting the devices from three-dimensional structures to two-dimensional structures.

The research community has remained divided in opinion as to what happens in the two-dimensional world of the domain walls: How is the border between two domains with different electric polarization stabilized? Is the polarization in domain walls different to the polarization in the domains themselves? Can the properties of the domain wall be controlled in a localized manner? The great interest in addressing these questions stems from the fact that a ferroelectric material in its natural form is an excellent electric insulator. However, the domain walls may be conducting electrically, thus forming a two-dimensional object that are controllable by will. This phenomenon encompasses the potential to reduce significantly the energy consumption of data storage and data processing devices.

In this project, the researchers succeeded in deciphering the atomic structure and electric field deployment in domain walls at the atomic scale. In their recent article, they corroborate the assumption that domain walls allow for the existence of a two-dimensional border between domains as a result of partial oxygen vacancy in areas that are common to two domains, thus enabling greater flexibility in the deployment of the local electric field. They succeeded in engineeringly inducing an individual oxygen atom vacancy and demonstrated that this action creates opposing dipoles and greater electric symmetry – a unique topological structure called a quadrupole.

With the aid of computer simulations by Shi Liu of Westlake University in China, the researchers demonstrated that engineering the oxygen atom vacancy has a great impact on the electrical properties of the material not only at the atomic scale, but also at the scale that is relevant to electronic devices – for example, in terms of electrical conductivity. The significance is that the present scientific achievement is likely to be of help in miniaturizing devices of this kind as well as reducing their energy consumption.

Collaboration with researchers from the Negev Nuclear Research Center, the Technion research group also demonstrated that oxygen vacancies can be engineered by exposing the material to electronic radiation. Consequently, in addition to the technological potential of the discovery in electronics, it may also be possible to utilize the effect for radiation detectors, allowing for the early detection – and prevention – of nuclear accidents, such as the one that happened in 2011 in Fukushima, Japan.

The research, which was carried out at the Electron Microscopy Center in the Faculty of Materials Science and Engineering, was funded by the Israel Science Foundation and the Pazy Foundation. The Nano and Quantum Functional Structures Laboratory, headed by Asst. Prof. Ivry, is supported by the Zuckerman STEM Leadership Program.

For the article in ACS Nano click here


Molecule that inhibits degenerative processes related to Alzheimer’s disease

Share on:

 Researchers at the Technion developed a molecule that inhibits degenerative processes related to Alzheimer’s disease

Professor Galia Maayan

Researchers at the Technion – Israel institute of Technology, in collaboration with researchers from CNRS, recently published findings about the development of an artificial molecule that may inhibit the development of Alzheimer’s disease. The molecule breaks down the toxic chemical complex Cu–Aβ, thus inhibiting the cell death that is thought to be related to Alzheimer’s. The study was led by Professor Galia Maayan and doctoral student Anastasia Behar from the Schulich Faculty of Chemistry, in collaboration with Prof. Christelle Hureau from the Laboratoire de Chimie de Coordination du CNRS, Toulouse, France.

Copper ions are a key component of the structure and function of various cells in the body. But their accumulation can lead to cell toxicity, causing dangerous conditions such as oxidative stress, cardiovascular disorders, and degenerative diseases of the brain, including Alzheimer’s.

Doctoral student Anastasia Behar

One of the mechanisms involved in the development of Alzheimer’s is the formation of free radicals that damage the brain cells. These are oxidizing agents formed, among other things, by Cu–Aβ, a complex of copper and amyloid beta. It is already known that the breakdown of this complex, and the removal of copper from the amyloid, prevents cell death, followed by the inhibition of the disease. The extraction of the copper is done by chelation – using molecules that bind the copper ions and extract them from the amyloid.

However, this is not a simple challenge, because the chelators must meet several critical chemical and kinetic conditions, including stability and resistance to oxidation-reduction reactions. It is also important that the chelator does not bind zinc ions during the copper extraction process, as they are also essential for neuron function (but do not cause toxicity when they are bound to the amyloid); if the chelator does not bind the zinc, it can continue to bind the copper ions, but if it binds zinc, copper binding will be inhibited.

In the figure, from left to right: Oxidation of copper ions in an amyloid complex (that also contains zinc ions) leads to the formation of a toxic amyloid complex and harmful oxidizing agents (ROS). The water-soluble chelator extracts the copper ion from the amyloid complex

The Technion and CNRS researchers report in the Angewandte Chemie on the successful development of a new artificial chelator that meets all these requirements. The chelator, called P3, is a peptide-like water-soluble synthetic molecule that performs its task selectively; it strongly binds copper and forms the complex CuP3, extracting the copper from the amyloid. By doing so, it inhibits and even suppresses the formation of harmful oxidizing agents, without creating new oxidation processes. Although it binds zinc ions and even extracts them from the amyloid faster than it extracts the copper ions, the binding to zinc is weaker, making the zinc-amyloid complex unstable, so in practice P3 mostly binds copper ions. by creating a new, stable complex, and inhibits the formation of harmful oxidizing agents (NO ROS), thereby neutralizing amyloid toxicity.

Click here for the paper in Angewandte Chemie


An Inflammation to Remember

Share on:

An Inflammation to Remember

Technion scientists uncover a physiological mechanism of psychosomatic illness

Your phone pings. It’s a notification from your friend, who you just went out for a drink with last night. According to her text, she has just tested positive for COVID-19. You start feeling your throat, you sneak a short cough, and start to feel your body temperature rising. But then you calm down (after receiving your negative COVID results, of course) and realize these feelings were all in your head. But what if this is exactly it – what if there were indeed neurons in the brain that could induce a sensation of illness, or even an actual disease?

Psychosomatic disorders are described as diseases emerging with no apparent biological cause, and often include a strong emotional component as a trigger. In a study recently published in Cell, Technion scientists explore the brain’s potential to cause diseases on its own. Specifically, they induced inflammation in mice, and then triggered the neurons in the brain that were active during the initial inflammation.

Tamar Koren (right) and Professor Asya Rolls

The study was conducted by the research group of Associate Professor Asya Rolls from the Technion Ruth and Bruce Rappaport Faculty of Medicine, led by Tamar Koren, an M.D./Ph.D. student in the lab. They showed that during colon inflammation, several brain regions exert enhanced neuronal activity, one of which was the insular cortex (insula). The insula is an area in the brain responsible for interoception, that is the sense of the body’s physiological state. This includes hunger, thirst, pain, and heart rate.

The researchers postulated that if report of inflammation in some area of the body is stored somewhere in the brain, this area responsible for interoception would be involved. Armed with this hypothesis, they induced in mice an inflammation in the colon and using genetic manipulation techniques, “captured” groups of neurons in the insular cortex that showed increased activity during the inflammation. Once the mice were healthy, the researchers triggered these “captured” neurons artificially. Without any outside stimulus other than this triggering of cells in the brain, inflammation re-emerged, in the exact same area where it was before. “Remembering” the inflammation was enough to reactivate it.

If the brain can generate disease, is it possible that it can also turn it off?

Scientific photo: Upper panel: Insular neurons (in red) that were captured during colitis and reactivated (in green) upon recovery. Lower panel: Colon sections showing white blood cells (in red) present in the tissue of a mouse after insular neurons reactivation (Gq, right) and its non-activated control (Sham, left).

In a similar manner, Tamar also demonstrated the opposite effect: in mice with active inflammation, suppressing the neurons that remembered it produced immediate reduction in the inflammation. Although this was a basic study in mice, and there are multiple challenges in translating the concept to humans, these discoveries open a new therapeutic avenue for treating chronic inflammatory conditions such as Crohn’s disease, psoriasis, and other autoimmune conditions, by attenuating their memory trace in the brain.

The Research group of Professor Asya Rolls

“There are evolutionary advantages to such a connection,” said Prof. Rolls in explaining the strange phenomenon whereby the immune system should be activated by memory alone, without an outside trigger. “The body needs to respond to infection as quickly as possible before the attacking bacteria or viruses can multiply. If certain activity, for example consuming particular foods, has exposed the body to infection and inflammation once, there is an advantage to gearing up for battle when one is about to engage in the same activity again. A shorter response time would allow the body to defeat the infection faster and with less effort. The problem of course is when such an effective mechanism goes out of control and can on its own generate the disease.”

The group’s findings have broad implications for understanding the way the human mind and body affect each other, but also more immediate implications for understanding and treating illness with a psychosomatic element, like irritable bowel syndrome, and even autoimmune diseases and allergies.

The study was done in collaboration with Dr. Kobi Rosenblum, of the University of Haifa and Dr. Fahed Hakim, of EMMS Hospital, Nazareth. This work was supported by the European Research Council (ERC) Starting Grant, the Allen and Jewel Prince Center for Neurodegenerative Disorders of the Brain, the Howard Hughes Medical Institute (HHMI), and the Wellcome trust.


For the full article in Cell click here.

Click here for video demonstrating the research

Scientists Discover Emergency Pathway to Help Human Cells with Protein Damage Survive

Share on:

Scientists Discover Emergency Pathway to Help Human Cells with Protein Damage Survive

Cell proteins damaged by oxygen radicals can be chemically “tagged” for elimination, but an “emergency pathway” bypasses strict protocol and can eliminate even without the need for prior tagging.

An international research team headed by Technion scientists has found an alternative manner for eliminating damaged proteins when the cells are impaired by “oxygen radicals,” as can happen in failing human hearts where there is poor cell respiration and cells become oxygen depleted, or suffer “hypoxia,” because of poor oxygen uptake.

Significantly, the researchers discovered that there can be a shift from the tightly controlled process of eliminating proteins in the cells to a less strict mechanism when cells enter an “emergency protocol.” This shift can “clear up” the toxic proteins before their toxicity levels get too high.

Their study was published on 26 October 2021 in Nature Communication. To carry out their study, the researchers investigated several “proteasomes,” protein complexes that work by a chemical reaction to degrade unneeded or damaged cell proteins. The researchers found that elevated levels of one type of proteasome, 20S, appears to contribute to cell survival, even for those cells under stress from damaged proteins.

L-R: Professor Oded Kleifeld, Professor Michael Glickman and Professor Ashraf Brik

Human cells – both functional and damaged – are constantly recycled by chemically “tagging” and targeting for removal when they are under stress by the ubiquitin system (2004 Nobel Prize in chemistry). At the same time, a few proteins that are intact and functional can also be dragged into the 20S proteasome “molecular disposal unit” along with the toxic proteins that have be targeted for destruction. Nevertheless, rather than harm cells, this mode of action by 20S proteasome may aid cells in rapidly remove toxic proteins. In their conclusion, the authors raised the interesting speculation that this emergency pathway can help even damaged cells to withstand bouts of stress and allow them to “age gracefully”.

Professor Michael Glickman (left) and Professor Indrajit Sahu

To carry out the study, Technion researchers Professors Indrajit Sahu, Michael Glickman, Ashraf Brik, and Oded Kleifeld, worked with Professor Sharlene Day, from the University of Pennsylvania, and the research team of Professor Yao Cong of the Chinese Academy of Sciences in Shanghai, China.


Click here for the paper in Nature Communications

“Nano-Taxis” Shuttle Therapeutics to Neurons

Share on:

“Nano-Taxis” Shuttle Therapeutics to Neurons

Technion – Israel Institute of Technology Assistant Professor Assaf Zinger and Dr. Caroline Cvetkovic from the Center for Neuroregeneration at the Houston Methodist Research Institute have created a novel means of delivering medicine to neurons in a targeted manner.

The biomimetic nano-vesicles, or nature-mimicking, nanometer-scale “vehicles,” are capable of specifically targeting neurons such as nerve cells. These nano-vehicles pave the way for the treatment of multiple neurodegenerative diseases and traumatic brain injuries.

Assistant Professor Assaf Zinger

Drug delivery is a major challenge that must be overcome in drug development, and it is one of the focus areas of the Wolfson Faculty of Chemical Engineering at the Technion. It is not enough that a substance can lead to the desired therapeutic effect in a specific cell. This therapeutic substance must also reach these cells without being changed or destroyed en route, and it must not end up in other organs if it might cause harm there.

Nano-vesicles are similar in structure to human cells, but much smaller — one millionth of a hair’s width in diameter. They can carry within them cargo that needs to be delivered to the cells such as medication, mRNA, etc.

Nano-vesicles can be targeted by incorporating specific cell membrane-derived proteins on their surface, thus letting them be recognized and taken in by the correct cells. In essence, the nano vesicles (or taxis) masquerade as neurons, resulting in their being recognized and welcomed by other neurons, thereby making it possible for them to deliver their therapeutic cargo.

Potentially revolutionizing the treatment of neurodegenerative disorders and traumatic brain injuries

“Neurosomes”- Humanized Biomimetic nano vesicles (red) for neuron targeting (green)

These findings have broad implications. More than one neurodegenerative disorder might be treated if the correct medicine or genetic cargo (e.g., mRNA, SiRNA, miRNA) could be delivered to the brain. But these are not the only possible applications.

“With [nano-taxis], we can also potentially revolutionize the treatment of traumatic brain injuries,” Prof. Zinger explained. “In the case of a car accident and or a sports injury, as examples, the brain is first damaged by the impact, as it is struck against the skull. As a result, multiple brain cells are damaged. This starts a process of inflammation. If we could immediately deliver anti-inflammatory drugs to the brain, we could reduce the inflammatory processes, and hopefully prevent fatalities and long-term disabilities.”

The lion’s share of this study was conducted by Prof. Zinger at the Houston Methodist Research Institute and Houston Methodist Hospital as part of his postdoctoral fellowship. Prof. Zinger recently opened a multidisciplinary laboratory at the Technion, in the Wolfson Faculty of Chemical Engineering. His lab aims to create advanced bioinspired technologies and translational therapeutics through a highly multidisciplinary approach.

Cancer Cells Mobilizing the Nervous System? Let’s Use Them to Inhibit the Tumor

Share on:

Cancer Cells Mobilizing the Nervous System? Let’s Use Them to Inhibit the Tumor

October is Breast Cancer Awareness Month, and Technion researchers have just published findings in Science Advances that support the efficacy of the technology that they developed: Treatment of breast cancer by anesthesia of the nervous system around the tumor. The treatment not only inhibited tumor growth but also prevented metastasis to other organs                                                                                                

Professor Avi Schroeder

Researchers at the Technion – Israel Institute of Technology have developed an innovative treatment for breast cancer, based on analgesic nanoparticles that target the nervous system. The study, published in Science Advances, was led by Professor Avi Schroeder and Ph.D. student Maya Kaduri of the Wolfson Faculty of Chemical Engineering.

Breast cancer is one of the most common cancers in women, and despite breakthroughs in diagnosis and treatment, about one thousand women in Israel die of the disease per year. Around 15% of them are under the age of 50. Worldwide, some 685,000 women die each year from breast cancer.

Prof. Schroeder has years of experience in developing innovative cancer treatments, including ones for breast cancer and specifically triple-negative breast cancer – an aggressive cancer characterized by rapid cell division with a higher risk of metastasis. Technologies developed in his lab include novel methods for encapsulating drug molecules in nanoparticles that transport the drug to the tumor and release it inside, without damaging healthy tissue.

The researchers found that cancer cells have a reciprocal relationship with the nerve cells around them: the cancer cells stimulate infiltration of nerve cells into the tumor, and this infiltration stimulates cancer cell proliferation, growth, and migration. In other words, the cancer cells recruit the nerve cells for their purposes.

Based on these findings, the researchers developed a treatment that targets the tumor through the nerve cells. This treatment is based on injecting nanoparticles containing anesthetic into the bloodstream. The nanoparticles travel through the bloodstream toward the tumor, accumulate around the nerve cells in the cancerous tissue, and paralyze the local nerves and communication between the nerve cells and the cancer cells. The result: significant inhibition of tumor development and of metastasis to the lungs, brain, and bone marrow.

The nanoparticles simulate the cell membrane and are coated with special polymers that disguise them from the immune system and enable a long circulation time in the bloodstream. Each such particle, which is around 100 nm in diameter, contains the anesthetic.

Maya Kaduri

According to Maya Kaduri: “We know how to create the exact size of particles needed, and that is critical because it’s the key to penetrating the tumor. Tumors stimulate increased formation of new blood vessels around them, so that they receive oxygen and nutrients, but the structure of these blood vessels is damaged and contains nano-sized holes that enable penetration of nanoparticles. The cancerous tissue is characterized by poor lymphatic drainage, which further increases accumulation of the particles in the tissue. Therefore, the anesthetizing particles we developed move through the bloodstream without penetrating healthy tissue. Only when they reach the damaged blood vessels of the tumor do they leak out, accumulate around the nerve cells of the cancerous tissue, and disconnect them from the cancer cells. The fact that this is a very focused and precise treatment enables us to insert significant amounts of anesthetic into the body because there is no fear that it will harm healthy and vital areas of the nervous system.”

In experiments on cancer cell cultures and in treatment of mice, the new technology inhibited not only tumor development but also metastasis. The researchers estimate these findings may be relevant for treatment of breast cancer in humans.

The research is supported by the Rappaport Technion Integrated Cancer Center (RTICC) as part of the Steven & Beverly Rubenstein Charitable Foundation Fellowship Fund for Cancer Research, and by Teva, as part of its National Forum for BioInnovators. The research was conducted in cooperation with the Faculty of Medicine at Hebrew University of Jerusalem and the Institute of Pathology at the Tel Aviv Sourasky Medical Center.

Prof. Avi Schroeder is head of the Louis Family Laboratory for Targeted Drug Delivery & Personalized Medicine Technologies at the Wolfson Faculty of Chemical Engineering. Maya Kaduri, who has a B.Sc. from the Faculty of Biotechnology and Food Engineering at the Technion, began researching under the guidance of Prof. Avi Schroeder during her bachelor’s degree, and this year she is expected to complete her Ph.D. (direct track).

For the article in Science Advances click here

Click here for video demonstrating the research


Machine Learning in Cardiology

Share on:

Machine Learning in Cardiology

Technion scientists use machine learning for atrial fibrillation risk prediction

Shany Biton

Shany Biton and Sheina Gendelman, two M.Sc. students working under the supervision of Assistant Professor Joachim A. Behar, head of the Artificial Intelligence in Medicine laboratory (AIMLab.) in the Technion Faculty of Biomedical Engineering, wrote a machine learning algorithm capable of accurately predicting whether a patient will develop atrial fibrillation within 5 years. Conceptually, the researchers sought to find out whether a machine learning algorithm could capture patterns predictive of atrial fibrillation even though there was no atrial fibrillation diagnosed by a human cardiologist at the time.

Assistant Professor Joachim Behar

Atrial fibrillation is an abnormal heart rhythm that is not immediately life-threatening, but significantly increases patients’ risk of stroke and death. Warning patients that they are at risk of developing it can give them time to change their lifestyle and avoid or postpone the onset of the condition. It may also encourage regular follow-ups with the patient’s cardiologist, ensuring that if and when the condition develops, it will be identified quickly, and treatment will be started without delay. Known risk factors for atrial fibrillation include sedentary lifestyle, obesity, smoking, genetic predisposition and more.

Sheina Gendelman

Ms. Biton and Ms. Gendelman used more than one million 12-lead ECG recordings from more than 400,000 patients to train a deep neural network to recognize patients at risk of developing atrial fibrillation within 5 years. Then, they combined the deep neural network with clinical information about the patient, including some of the known risk factors. Both the ECG recordings and the patients’ electronic health record were provided by the Telehealth Network of Minas Gerais (TNMG), a public telehealth system assisting 811 of the 853 municipalities in the state of Minas Gerais, Brazil. The resulting machine learning model was able to correctly predict the development of atrial fibrillation risk in 60% of cases, while preserving a high specificity of 95%, meaning that only 5% of persons identified as being potentially at risk did not develop the condition.

“We do not seek to replace the human doctor – we don’t think that would be desirable,” said Prof. Behar of the results, “but we wish to put better decision support tools into the doctors’ hands. Computers are better equipped to process some forms of data. For example, examining an ECG recording today, a cardiologist would be looking for specific features which are known to be associated with a particular disease. Our model, on the other hand, can look for and identify patterns on its own, including patterns that might not be intelligible to the human eye.”

Overview of the experimental setting: digital biomarkers (HRV and MOR), deep learning features (DNN) and clinical data (EMR) are combined together in training a model to predict the future occurrence of atrial fibrillation

Doctors have progressed from taking a patient’s pulse manually, to using a statoscope, and then the ECG. Using machine learning to assist the analysis of ECG recordings could be the next step on that road.

Since ECG is a low-cost routine test, the machine learning model could easily be incorporated into clinical practice and improve healthcare management for many individuals. Access to more patients’ datasets would let the algorithm get progressively better as a risk prediction tool. The model could also be adapted to predict other cardiovascular conditions.

The study was conducted in collaboration with Antônio Ribeiro from the Uppsala University, Sweden and Gabriela Miana, Carla Moreira, Antonio Luiz Ribeiro from the Universidade Federal de Minas Gerais, Brazil.

The study was published in the European Heart Journal – Digital Health.


This E-Skin Knows What Movement You Make

Share on:


This E-Skin Knows What Movement You Make

Technion scientists created a wearable motion sensor capable of identifying bending and twisting

The new device


Professor Hossam Haick

Technion scientists have produced a highly stretchable electronic material and created a wearable sensor capable of precisely identifying bending and twisting motion. It is essentially an electronic skin capable of recognizing the range of movement human joints normally make, with up to half a degree precision. This breakthrough is the result of collaborative work between researchers from different fields in the Laboratory for Nanomaterial-Based Devices, headed by Professor Hossam Haick from the Wolfson Faculty of Chemical Engineering. It was recently published in Advanced Materials and was featured on the journal’s cover.

Yehu David Horev

Prof. Haick’s lab is focused on wearable devices for various uses. Currently existing wearable motion sensors can recognize bending movement, but not twisting. Existing twisting sensors, on the other hand, are large and cumbersome. This problem was overcome by Ph.D. candidate Yehu David Horev and postdoctoral fellow Dr. Arnab Maity. Mr. Horev found a way to form a composite material that was both conductive (and thus, usable as a sensor) and flexible, stretchable, breathable, and biocompatible, and that did not change its electrical properties when stretched. Dr. Maity then solved the mathematics of analyzing the received signal, creating an algorithm capable of mapping bending and twisting motion – the nature of the movement, its speed, and its angle. The novel sensor is breathable, durable, and lightweight, allowing it to be worn on the human body for prolonged periods.

“This sensor has many possible applications,” Prof. Haick stated. “It can be used in early disease diagnosis, alerting of breathing alterations, and motor system disorders such as Parkinson’s disease. It can be used to assist patients’ motor recovery and be integrated into prosthetic limbs. In robotics, the feedback it provides is crucial for precise motion. In industrial uses, such sensors are necessary in monitoring systems, putting them at the core of the fourth industrial revolution.”

“Electrically conductive polymers are usually quite brittle,” explained Mr. Yehu about the challenge the group had overcome. “To solve this, we created a composite material that is a little like fabric: the individual polymer ‘threads’ cannot withstand the strain on the material, but their movement relative to each other lets it stretch without breaking. It is not too different from what lends stretch to t-shirts. This allows the conductive polymer withstand extreme mechanical conditions without losing its electrical properties.”

What makes this achievement more important is that the materials the group used are very cheap, resulting in an inexpensive sensor. “If we make a device that is very expensive, only a small number of institutions in the Western World can afford to use it. We want the technological advances we achieve to benefit everyone, regardless of their geographic location and socio-economic status,” said Prof. Haick. True to his word, among the laboratory’s other projects is a tuberculosis-diagnosing sticker patch, sorely needed in developing countries.

Dr. Arnab Maity

The scientists who contributed to this study are Yehu David Horev, Dr. Arnab Maity, Dr. Youbin Zheng, Yana Milyutin, Dr. Muhammad Khatib, Dr. Ning Tang, and Prof. Hossam Haick from the Department of Chemical Engineering and Russell Berrie Nanotechnology Institute at the Technion-Israel Institute of Technology; Miaomiao Yuan from the Eighth Affiliated Hospital, Sun Yat-sen University, China; Dr. Ran Yosef Suckeveriene from the Department of Water Industry Engineering at the Kinneret Academic College; and Prof. Weiwei Wu from the School of Advanced Materials and Nanotechnology at Xidian University, China.

Click here for the paper in Advanced Materials