Big data and machine learning: new opportunities for medicine
“We have developed more advanced technologies for shoe selection on Amazon than for choosing the type of treatment for cancer patients,” says MIT professor Regina Barzilay about the current state of high-tech medical projects. The assessment is disappointing: the often “popular” areas, such as e-commerce, are ahead of more socially important areas in terms of the level of technologies used.
However, there is good news: solutions that were developed for the conditional “shoe search” can also be used to help patients. And the demand for such developments is only growing: according to Frost & Sullivan, the volume of the medical development market using machine learning and big data alone is increasing by 40% annually and by 2021 will amount to 6.6 billion dollars.
Today we will talk about how big data is used in medical projects and what developments in this direction are carried out at ITMO University.
')
Photo by Charles Clegg CC-BY
Data mining, machine learning and processing of natural language, in particular, are actively used to solve problems of early diagnosis of diseases: from cancer and diabetes to schizophrenia. For example, the American project PathAI does an excellent job of detecting breast cancer in its early stages. In April 2016, the system competed with an expert and lost: his error rate was 3.5%, and that of the system - 7.5%. Since then, researchers have been able to increase the sample size at which the system was trained, and by November of that year, PathAI had surpassed the expert in diagnostic accuracy.
As Joel Dudley, developer of the Deep Patient System at Mount Sinai Hospital in New York, says : “One of the important features of in-depth training is that when generating forecasts or models, you do not need to limit yourself in advance to only the most important information.” . This applies, for example, to the analysis of the entire patient's medical history during the formation of a treatment plan. Or comparing individual patient data with information about other cases — the Deep Patient algorithm accesses a database of 5 million people.
However, the use of big data in medicine is not limited to these examples. For example, at ITMO University, one of the projects, combining big data and medicine, is implemented by the Institute for High-Tech Computer Technologies. Together with the Northwestern Federal Medical Center named after V.A. Almazov is developing a fleet management system for ambulances in St. Petersburg. The objective of the project is to help dispatchers organize the most expeditious hospitalization.
In solving this problem, the system takes into account the statistics of calls for emergency assistance, data on the mobility of the population during the day, data on the load of transport networks and emergency departments of hospitals. As a result, the solution allows, firstly, to optimize the routes of the ambulance, and secondly, it helps to formulate recommendations for improving the operating procedures of ambulance stations.
Development will develop in two directions: on the one hand, the decision support system will accumulate information on an increasing number of diseases. On the other hand, the project will be supplemented with a solution for automating medical record keeping.
By the way, predictive modeling and working with big data in medicine is not just a special case of using technology, but an independent scientific field. ITMO University specialists are trained in it at the department of high-performance computing in the framework of the master's program "Computational Biomedicine" .
Undergraduates study the methods, algorithms and technologies used in bioinformatics, genomic and epidemiological studies, when creating drugs. In addition, the training course includes the study of models of physiological processes in the human body, as well as the processes of health care institutions and other basic knowledge, which allow an IT specialist and analyst to talk to doctors, biologists, and chemists in the same language.
Speaking of chemistry: another area for working with big data in the medical sector is biological and chemical research and a related discipline — chemo informatics. When creating a new compound, for example, for a medical drug, it is necessary to conduct a lot of experiments and tests. Chemoinformatics allows you to speed up this process due to its modeling based on modern databases and machine learning algorithms.
By the way, the development of this discipline itself and especially the use of big data seriously changed the medical industry as a whole. The need to create very large amounts of data leads to the fact that pharmaceutical companies join forces and work together with independent scientific and research centers - ten years ago this practice seemed unlikely.
The “side effect” of working with big data in this area is the possibility of accumulating sufficient amounts of information to study so-called neglected diseases - diseases that are common among the poorest and most marginal groups of people living mainly in Asia, Africa and Latin America. Drug development and the study of these diseases are considered to be economically disadvantageous (for pharmaceutical companies). However, access to large data and, in particular, the emergence of open databases of chemical compounds and reactions can seriously reduce the cost of the process and give groups of enthusiasts the opportunity to work independently on solving such problems, at least, without the initial support of large pharmaceutical corporations.
At ITMO University, it is possible to study this area and work on your own project as part of the master's program “Chemoinformatics and Molecular Modeling” , which is conducted jointly with the University of Strasbourg. Future masters are learning to use (and develop) methods for constructing and analyzing databases of chemical compounds and reactions in order to predict their chemical and biological properties, predict the course of reactions, and solve the problems of finding new drugs.
However, there is good news: solutions that were developed for the conditional “shoe search” can also be used to help patients. And the demand for such developments is only growing: according to Frost & Sullivan, the volume of the medical development market using machine learning and big data alone is increasing by 40% annually and by 2021 will amount to 6.6 billion dollars.
Today we will talk about how big data is used in medical projects and what developments in this direction are carried out at ITMO University.
')
Photo by Charles Clegg CC-BYDiagnosis of diseases
Data mining, machine learning and processing of natural language, in particular, are actively used to solve problems of early diagnosis of diseases: from cancer and diabetes to schizophrenia. For example, the American project PathAI does an excellent job of detecting breast cancer in its early stages. In April 2016, the system competed with an expert and lost: his error rate was 3.5%, and that of the system - 7.5%. Since then, researchers have been able to increase the sample size at which the system was trained, and by November of that year, PathAI had surpassed the expert in diagnostic accuracy.
As Joel Dudley, developer of the Deep Patient System at Mount Sinai Hospital in New York, says : “One of the important features of in-depth training is that when generating forecasts or models, you do not need to limit yourself in advance to only the most important information.” . This applies, for example, to the analysis of the entire patient's medical history during the formation of a treatment plan. Or comparing individual patient data with information about other cases — the Deep Patient algorithm accesses a database of 5 million people.
Simulation of work "ambulances"
However, the use of big data in medicine is not limited to these examples. For example, at ITMO University, one of the projects, combining big data and medicine, is implemented by the Institute for High-Tech Computer Technologies. Together with the Northwestern Federal Medical Center named after V.A. Almazov is developing a fleet management system for ambulances in St. Petersburg. The objective of the project is to help dispatchers organize the most expeditious hospitalization.
In solving this problem, the system takes into account the statistics of calls for emergency assistance, data on the mobility of the population during the day, data on the load of transport networks and emergency departments of hospitals. As a result, the solution allows, firstly, to optimize the routes of the ambulance, and secondly, it helps to formulate recommendations for improving the operating procedures of ambulance stations.
Development will develop in two directions: on the one hand, the decision support system will accumulate information on an increasing number of diseases. On the other hand, the project will be supplemented with a solution for automating medical record keeping.
Computational Biomedicine
By the way, predictive modeling and working with big data in medicine is not just a special case of using technology, but an independent scientific field. ITMO University specialists are trained in it at the department of high-performance computing in the framework of the master's program "Computational Biomedicine" .
Undergraduates study the methods, algorithms and technologies used in bioinformatics, genomic and epidemiological studies, when creating drugs. In addition, the training course includes the study of models of physiological processes in the human body, as well as the processes of health care institutions and other basic knowledge, which allow an IT specialist and analyst to talk to doctors, biologists, and chemists in the same language.
Chemoinformatics
Speaking of chemistry: another area for working with big data in the medical sector is biological and chemical research and a related discipline — chemo informatics. When creating a new compound, for example, for a medical drug, it is necessary to conduct a lot of experiments and tests. Chemoinformatics allows you to speed up this process due to its modeling based on modern databases and machine learning algorithms.
By the way, the development of this discipline itself and especially the use of big data seriously changed the medical industry as a whole. The need to create very large amounts of data leads to the fact that pharmaceutical companies join forces and work together with independent scientific and research centers - ten years ago this practice seemed unlikely.
The “side effect” of working with big data in this area is the possibility of accumulating sufficient amounts of information to study so-called neglected diseases - diseases that are common among the poorest and most marginal groups of people living mainly in Asia, Africa and Latin America. Drug development and the study of these diseases are considered to be economically disadvantageous (for pharmaceutical companies). However, access to large data and, in particular, the emergence of open databases of chemical compounds and reactions can seriously reduce the cost of the process and give groups of enthusiasts the opportunity to work independently on solving such problems, at least, without the initial support of large pharmaceutical corporations.
At ITMO University, it is possible to study this area and work on your own project as part of the master's program “Chemoinformatics and Molecular Modeling” , which is conducted jointly with the University of Strasbourg. Future masters are learning to use (and develop) methods for constructing and analyzing databases of chemical compounds and reactions in order to predict their chemical and biological properties, predict the course of reactions, and solve the problems of finding new drugs.
Source: https://habr.com/ru/post/340668/
All Articles