By Gabriel Lobitsky, sales director at Infor
With data analysis it is possible to teach the machine to act proactively to support preventive medicine
Machine Learning everywhere! Not only in industry, where its application has been common, but in areas quite dependent on human intervention, such as finance, logistics and health. Specifically for the latter, the advanced techniques of data analysis and machine learning bring an exponential gain: they allow to look ahead and assist preventive medicine. You can imagine how it changes the business landscape of health insurance companies, offices and hospitals, if based on patient data, it was possible to use data analysis in a more intelligent way to suggest preventive care at certain times of the year, or even promote a healthier lifestyle?
In a practical way, machine learning is used to model the algorithms and give intelligent answers from advanced data analysis, so big data can contain a lot of information and the answers can be in a little data. The repertoire of information that scientists need can be found in software and medical equipment that emit data, and to achieve the ideal data model, it is necessary to model the algorithm, refine it, so that the machine learns - with the help of human beings - to achieve an efficiency model and become more intelligent and proactive.
So, imagine the following situation: a health insurance company can mine the data records that hospitals issue about their patients through a CRM (Customer Relationship Management). From the selected information, it is possible to create campaigns and preventive actions for a specific group of customers. These customers may be employees of the same company that had dengue in the summer. The health insurance plan can proactively suggest preventive and informative action on the disease to prevent new cases.
So there is no doubt: machine learning can save lives! This year, the University of Florida published a survey with this bias - the impact of which is such that it has become one of the most talked about studies around the world. You may have seen it, but it is worth remembering that researchers at the institution extracted data related to patients who have tried to commit suicide, from electronic medical records, and used ML techniques to identify groups of people with a tendency to try something against their own lives. . The machine learning algorithms developed by them, are capable of predicting suicide attempts with up to 90% accuracy, up to two years before. Fantastic, isn't it?
What step are we in?
Today, in Brazil, the use of machine learning is more applied in recommendations for consumers such as music, movies, travel, product and service offers. However, it is necessary to look at this model in a more strategic way, because Brazil still faces numerous problems in the area of health that technology alone cannot solve. They are old equipment that do not communicate or emit information; and for not having interoperability with software and other systems, they prevent machine learning from being used in its entirety.
With increasingly accurate mathematical algorithms and based on this history of data, the machine is able to alert about possible diseases, identify trend groups, among other factors. Machine learning can be applied by linking each patient record, in the different data sets, automatically, to build a more complete picture of the activity.
The data here is the driving force. With it, healthcare companies can become smarter, reinvent their processes, create new business and service models. And how can health care prepare for all of this? Defining and planning the data: where are they (medical records, wearables, mobile apps)? In general, large companies already have their CIOs looking at this. And the ones that don't? I strongly recommend that you do so!
Machine Learning everywhere! Not only in industry, where its application has been common, but in areas quite dependent on human intervention, such as finance, logistics and health. Specifically for the latter, the advanced techniques of data analysis and machine learning bring an exponential gain: they allow to look ahead and assist preventive medicine. You can imagine how it changes the business landscape of health insurance companies, offices and hospitals, if based on patient data, it was possible to use data analysis in a more intelligent way to suggest preventive care at certain times of the year, or even promote a healthier lifestyle?
In a practical way, machine learning is used to model the algorithms and give intelligent answers from advanced data analysis, so big data can contain a lot of information and the answers can be in a little data. The repertoire of information that scientists need can be found in software and medical equipment that emit data, and to achieve the ideal data model, it is necessary to model the algorithm, refine it, so that the machine learns - with the help of human beings - to achieve an efficiency model and become more intelligent and proactive.
So, imagine the following situation: a health insurance company can mine the data records that hospitals issue about their patients through a CRM (Customer Relationship Management). From the selected information, it is possible to create campaigns and preventive actions for a specific group of customers. These customers may be employees of the same company that had dengue in the summer. The health insurance plan can proactively suggest preventive and informative action on the disease to prevent new cases.
So there is no doubt: machine learning can save lives! This year, the University of Florida published a survey with this bias - the impact of which is such that it has become one of the most talked about studies around the world. You may have seen it, but it is worth remembering that researchers at the institution extracted data related to patients who have tried to commit suicide, from electronic medical records, and used ML techniques to identify groups of people with a tendency to try something against their own lives. . The machine learning algorithms developed by them, are capable of predicting suicide attempts with up to 90% accuracy, up to two years before. Fantastic, isn't it?
What step are we in?
Today, in Brazil, the use of machine learning is more applied in recommendations for consumers such as music, movies, travel, product and service offers. However, it is necessary to look at this model in a more strategic way, because Brazil still faces numerous problems in the area of health that technology alone cannot solve. They are old equipment that do not communicate or emit information; and for not having interoperability with software and other systems, they prevent machine learning from being used in its entirety.
With increasingly accurate mathematical algorithms and based on this history of data, the machine is able to alert about possible diseases, identify trend groups, among other factors. Machine learning can be applied by linking each patient record, in the different data sets, automatically, to build a more complete picture of the activity.
The data here is the driving force. With it, healthcare companies can become smarter, reinvent their processes, create new business and service models. And how can health care prepare for all of this? Defining and planning the data: where are they (medical records, wearables, mobile apps)? In general, large companies already have their CIOs looking at this. And the ones that don't? I strongly recommend that you do so!
EN 
PT 
