Analysing Genetic Information Can Help Prevent Complications after Myocardial Infarction
Researchers at HSE University have developed a machine learning (ML) model capable of predicting the risk of complications—major adverse cardiac events—in patients following a myocardial infarction. For the first time, the model incorporates genetic data, enabling a more accurate assessment of the risk of long-term complications. The study has been published in Frontiers in Medicine.
Coronary artery disease (CAD), or ischaemic heart disease (IHD), is a condition characterised by insufficient blood and oxygen supply to the heart from narrowing or blockage of the coronary arteries. It is typically caused by plaques composed of fats and cholesterol that build up on the walls of blood vessels. Coronary heart disease may present as angina (chest pain), myocardial infarction (heart attack), or other problems.
According to WHO, ischaemic heart disease is the world’s biggest killer, responsible for 13% of the total deaths. Therefore, it is crucial to prescribe appropriate treatment to minimise the risks of complications and recurrences. Researchers at HSE University developed a model capable of predicting the probability of major adverse cardiac events following a myocardial infarction.
The scientists analysed data from patients admitted with myocardial infarction to the Surgut District Centre for Diagnostics and Cardiovascular Surgery between 2015 and 2024. Upon admission to the emergency department, medical researchers (cardiologists) explained the main points of the study protocol to the patients and obtained their informed consent to participate. The cardiologists then assessed the condition of the coronary arteries supplying the heart and based on their evaluation, either balloon angioplasty with stenting or coronary artery bypass grafting were performed. All patients received guideline-based therapy, including RAAS-blockers, beta-blockers, statins, and dual antiplatelet therapy. The information was documented in the patients' hospital medical records. For each patient, standard clinical measurements were taken, including blood pressure, body mass index, and cholesterol and glucose levels.
During the laboratory stage, DNA was isolated from the leukocyte rings in the collected blood samples and then frozen at −80°C for future genetic testing. The genotypes were determined based on a specific genetic variation (polymorphism) in the VEGFR-2 gene. The genetic marker VEGFR-2 is a component of the body's signalling system that regulates the growth of new blood vessels. There are three variations of the genotype—C/C, C/T, and T/T—differing in the variation of the DNA nucleotides cytosine (C) or thymine (T) in this region of the gene. Although the marker has been known for a long time, this was the first study to examine its impact on the prognosis of complications following myocardial infarction.
The authors evaluated the impact of 39 factors on the prognosis of risks such as cardiac death, recurrent acute coronary syndrome, stroke, and the need for repeat revascularisation, a procedure that helps restore blood flow in the arteries. To select the best model, the researchers trained and tested several machine learning algorithms: Gradient Boosting (CatBoost and LightGBM), Random Forest, Logistic Regression, and an AutoML approach.
The CatBoost model, a gradient boosting algorithm optimised for working with categorical data rather than numeric values, demonstrated the best performance. It makes predictions by sequentially building and training 'weak' decision trees, where each new tree corrects the errors of the previous ones. When building trees, the algorithm splits the data into two parts: the model is trained on one portion, while errors are calculated on the other. This reduces the risk of overfitting, where the model simply memorises the correct answers, and helps it identify general patterns for making predictions in new, unseen cases.
The influence of features on the model's accuracy was evaluated using the method of sequential feature addition, which assesses their contribution at each stage. The researchers selected the 9 most significant features: gender, body mass index, Charlson comorbidity index (which accounts for the presence of serious concomitant diseases), condition of the lateral wall of the left ventricle, degree of damage to the left coronary artery trunk, number of affected arteries, variant of the VEGFR-2 gene, choice between percutaneous coronary intervention or coronary artery bypass grafting, and statin dosage.
The results showed that the dose of statins, medications used to lower cholesterol levels in the blood, is the most important factor influencing the risk of complications. High doses of statins reduce this risk, particularly in patients with an unfavourable genotype. The VEGFR-2 polymorphism, specifically the presence of the T allele, was ranked fourth in terms of importance.
'Previously, genetic factors were not included in ML models, primarily because sequencing or even genotyping of individual nucleotides is not routinely performed in hospitals. In addition to standard measurements, we had access to data on polymorphism in the VEGFR-2 gene. This allowed us to compare this indicator with others and determine that the risk allele of the VEGFR-2 variant is one of the five most important factors for predicting long-term outcomes in patients with myocardial infarction,' explains Maria Poptsova, Head of the HSE International Laboratory of Bioinformatics and co-author of the paper.
The researchers emphasise that analysing genetic data contributes to creating more accurate and personalised models for predicting the risk of major adverse cardiovascular events in patients following a myocardial infarction.
'Cardiovascular diseases require resources for diagnosis, treatment, rehabilitation, and prevention, and therefore place a significant burden on the healthcare system. The introduction of such models into clinical practice could reduce mortality and the frequency of recurrent infarctions, optimise treatment, and ease the burden on healthcare professionals,' according to Alexander Kirdeev, Research Assistant at the International Laboratory of Bioinformatics and lead author of the paper.
The study was carried out in the framework of HSE University's 'Mirror Laboratories' project.
See also:
Russian Scientists Develop New Compound for Treating Aggressive Tumours
A team of Russian researchers has synthesised a novel compound for boron neutron capture therapy (BNCT), a treatment for advanced cancer that uses the boron-10 isotope. The compound exhibits low toxicity, excellent water solubility, and eliminates the need for administering large volumes. Most importantly, the active substance reaches the tumour with minimal impact on healthy tissues. The study was published in the International Journal of Molecular Sciences shortly before World Cancer Day, observed annually on February 4.
Scientists Discover Link Between Brain's Structural Features and Autistic Traits in Children
Scientists have discovered significant structural differences in the brain's pathways, tracts, and thalamus between children with autism and their neurotypical peers, despite finding no functional differences. The most significant alterations were found in the pathways connecting the thalamus—the brain's sensory information processing centre—to the temporal lobe. Moreover, the severity of these alterations positively correlated with the intensity of the child's autistic traits. The study findings have been published in Behavioural Brain Research.
Earnings Inequality Declining in Russia
Earnings inequality in Russia has nearly halved over the past 25 years. The primary factors driving this trend are rising minimum wages, regional economic convergence, and shifts in the returns on education. Since 2019, a new phase of this process has been observed, with inequality continuing to decline but driven by entirely different mechanisms. These are the findings made by Anna Lukyanova, Assistant Professor at the HSE Faculty of Economic Sciences, in her new study. The results have been published in the Journal of the New Economic Association.
Russian Physicists Discover Method to Increase Number of Atoms in Quantum Sensors
Physicists from the Institute of Spectroscopy of the Russian Academy of Sciences and HSE University have successfully trapped rubidium-87 atoms for over four seconds. Their method can help improve the accuracy of quantum sensors, where both the number of trapped atoms and the trapping time are crucial. Such quantum systems are used to study dark matter, refine navigation systems, and aid in mineral exploration. The study findings have been published in the Journal of Experimental and Theoretical Physics Letters.
HSE Scientists Develop Application for Diagnosing Aphasia
Specialists at the HSE Centre for Language and Brain have developed an application for diagnosing language disorders (aphasia), which can result from head injuries, strokes, or other neurological conditions. AutoRAT is the first standardised digital tool in Russia for assessing the presence and severity of language disorders. The application is available on RuStore and can be used on mobile and tablet devices running the Android operating system.
HSE Researchers Discover Simple and Reliable Way to Understand How People Perceive Taste
A team of scientists from the HSE Centre for Cognition & Decision Making has studied how food flavours affect brain activity, facial muscles, and emotions. Using near-infrared spectroscopy (fNIRS), they demonstrated that pleasant food activates brain areas associated with positive emotions, while neutral food stimulates regions linked to negative emotions and avoidance. This approach offers a simpler way to predict the market success of products and study eating disorders. The study was published in the journal Food Quality and Preference.
Russian Scientists Demonstrate How Disorder Contributes to Emergence of Unusual Superconductivity
Researchers at HSE University and MIPT have investigated how the composition of electrons in a superconductor influences the emergence of intertype superconductivity—a unique state in which superconductors display unusual properties. It was previously believed that intertype superconductivity occurs only in materials with minimal impurities. However, the scientists discovered that the region of intertype superconductivity not only persists but can also expand in materials with a high concentration of impurities and defects. In the future, these superconductors could contribute to the development of highly sensitive sensors and detectors. The study has been published in Frontiers of Physics.
HSE Scientists Take Important Step Forward in Development of 6G Communication Technologies
Researchers at HSE MIEM have successfully demonstrated the effective operation of a 6G wireless communication channel at sub-THz frequencies. The device transmits data at 12 Gbps and maintains signal stability by automatically switching when blocked. These metrics comply with international 6G standards. An article published on arXiv, an open-access electronic repository, provides a description of certain elements of the system.
AI vs AI: Scientists Develop Neural Networks to Detect Generated Text Insertions
A research team, including Alexander Shirnin from HSE University, has developed two models designed to detect AI-generated insertions in scientific texts. The AIpom system integrates two types of models: a decoder and an encoder. The Papilusion system is designed to detect modifications through synonyms and summarisation by neural networks, using one type of models: encoders. In the future, these models will assist in verifying the originality and credibility of scientific publications. Articles describing the Papilusion and AIpom systems have been published in the ACL Anthology Digital Archive.
Acoustic Battles for the Harem: How the Calls of Siberian Wapiti Reveal Their Status and Individuality
Researchers at HSE University, Lomonosov Moscow State University, and the A.N. Severtsov Institute of Ecology and Evolution of the Russian Academy of Sciences have studied the distinctive vocalisations of Siberian wapiti (Cervus canadensis sibiricus) stags during the peak of the mating season, when males produce rutting calls (bugles) to attract females (hinds) and deter rivals. The scientists have discovered how the acoustic parameters of these rutting calls reflect the stag's status—whether he currently holds a harem or is still attempting to acquire one—as well as his individual characteristics. The study has been published in Journal of Zoology.