Published in

MLearning.ai

# When Mathematics Meets Bio-Science: Data Science Implication

It was a healthy discussion with my principal while selecting between Mathematics and Biology post matriculation. Mathematics was always in my nerve with the spices of computer application in those school days it adds much more flavour, though I selected Biology, as I felt how much it is fascinated to know about what’s inside your body. Now, today I can relate both aspects if you deep dive into the human body, it will provide you with a large volume of data.

In this blog, I will be discussing the role of data science in the human body in the following areas:

1. Human Physiology

In layman terms, physiology means all the good things happening inside the body. In the recent decade, the interdisciplinary field of applied mathematical modelling in human physiology has grown enormously and continues to grow. The increasing ability of researchers to obtain data is one of the causes behind this growth. Because of faster sampling technologies and better means for getting both invasive and noninvasive data, the amount of physiological data received from diverse investigations is expanding tremendously.

Furthermore, data has a far higher time and space resolution than it did just a few years ago. For example, noninvasive magnetic resonance imaging (MRI) measurements can provide information on blood velocity as a function of time and three spatial coordinates in the heart and arteries as small as a few millimetres in diameter. Another recent achievement is the capacity to picture neuronal activity in the brain by observing variations in capillary oxygen levels.

Because of the vast amount of data, the models can provide both qualitative and quantitative information about the function they predict, as well as suggest additional trials. I believe that such models are required for a better understanding of the underlying physiology’s function and that in the long run, mathematical models may aid in the generation of new mathematical and physiological theories.

The following are some examples:

• Modelling the time delay associated with the baroreceptor system may provide insight into what causes Mayer waves (specific oscillations in mean arterial pressure).
• The presence of the dicrotic notch in the pulse profile and how the dicrotic notch changes throughout the aorta may be explained by modelling the propagation of the pulse wave down the aorta.
• Modelling the dynamics of cerebral blood flow response to abrupt hypotension after a shift in position from sitting to standing could help us better understand cerebral autoregulation.

Furthermore, models can aid in the avoidance of ambiguity, misconceptions, and lost effort. Only the use of mathematics can clearly define most, if not all, notions. Obscurity and ambiguity will occur sooner or later if mathematical descriptions are not provided. For example, when a few scholars offered distinct single indices to characterise the contractile condition of the ventricle, there was ambiguity. Some of these indicators are heavily reliant on the circulatory system; hence, rather than describing the ventricle’s contractile state, they describe the interplay between the ventricle and the vascular system.

Henceforth, if with the means of sophisticated technology that helps in the collection of data with ‘ n ’ number of attributes (it could be either labelled data or not) from the human body, it could help in demystifying the various question and helps in medical science.

2. Biomechanics

In simpler terms, Biomechanics is the study of forces and angles concerning the human body in rest or motion. Kinetics (when we are talking about forces) and Kinematics (when we are talking about angles) are two pillars of biomechanics. It helps in identifying force distribution while doing any sorts of activities (walking, running or playing sports) with the influence of angle. It is vital to understand the pathway of forces because not only does it prevent getting injured but also provide optimal performance from that particular muscle. For example, while throwing is one continuous, fluid motion, it can be broken down into six distinct phases. These segments include the: wind-up (variably present as this will be absent if the athlete is throwing from the stretch), early cocking, late cocking, acceleration, deceleration, and follow-through.

Integration of kinetics and kinematics in each phase could help to identify tricky movement that results in decrement of force and efficiency of the motion goes down, thus if we can sort out the faulty kinematics (intrinsic and extrinsic), it helps in the generation of optimal forces (with subject to human, ground reaction force, muscle force, joint force, gravitational forces).

With the collection of these data, we can do a supervised learning model having label data such as good biomechanics and faulty biomechanics, which helps in the understanding of loopholes causing issues like pain, decreased performance, et cetra.

3. Injury/ Disease Prediction

It is difficult to identify injury and disease as many factors contribute whether by injury or disease, this section of the blog is divided into two components:

• Injury Prediction — Injury could happen to anyone, it is that thing we can’t control if it is employing an accident more precisely collision injuries. However, when we deep dive into types of injuries it is categorized into two types: traumatic injuries and non-traumatic injuries. Non-traumatic injuries are overuse injuries, in which we have control and prevent it from happening.

Ask a sportsmen about injury they will give you a better narrative.

An injury could result in hampering the performance, time and money, and as there is quote:

Prevention is better than cure…!!

We can prevent the injury with the use of data science, you will be thinking how?

Injury happens with an asymmetry occurs in anatomical area and a certain load put in the tissue, in case of traumatic injury load is sudden cause breakage to tissue, however, in case of overuse injuries (atraumatic) load increase gradually causing wear and tear periodically.

If you can monitor the asymmetry and load, we can prevent the injury. Asymmetry could be monitor with the implication of biomechanics as discussed in earlier section.

Load Monitoring plays vital role in sports that helps in clustering athlete into:

a) Undertraining — it also leads to an injury

b) Optimal — probability of injuries are very minimal

c) Overtraining — chances of injuries are more

• Disease — Several contributing risk factors, such as diabetes, high blood pressure, excessive cholesterol, irregular pulse rate, and others, make it difficult to diagnose disease. The severity of sickness in people has been determined using a variety of data mining and neural network techniques. Various methods, such as the K-Nearest Neighbor Algorithm (KNN), Decision Trees (DT), Genetic Algorithm (GA), and Naive Bayes, are used to classify the severity of the condition (NB).

For example, in cardiac disease; because the nature of cardiac disease is complex, it must be treated with caution. Failure to do so may harm the heart or result in premature death. Medical science and data mining are utilised to uncover different types of metabolic disorders. In the prediction of heart disease and data inquiry, data mining with classification plays an important role.

The identification of raw healthcare data of cardiac information processing will aid in the long-term saving of human lives and the early detection of irregularities in heart problems.

However, if the condition is recognised early and preventative measures are taken as soon as feasible, the fatality rate can be dramatically reduced.

4. Bioinformatics

Bioinformatics is a branch of data science that focuses on using software (such as BLAST and Ensembl) to analyse biological data at the genomic and protein levels. Bioinformatics discoveries can benefit health care, agriculture, and biodiversity. It’s a combination of biology, computer science, statistics, and mathematics, which aren’t typically studied together. Typically, a specialist from one of the disciplines decides to pursue bioinformatics, which necessitates them becoming acquainted with the remaining fields.

Sequence analysis of DNA and proteins using various programmes and databases available on the internet is a vital activity. Anyone with access to the internet and relevant websites, from doctors to molecular biologists, can now use simple bioinformatic methods to uncover the composition of biological molecules such as nucleic acids and proteins. This isn’t to say that handling and analysing raw genetic data is simple for everyone. Expert bioinformaticians now use complex software programmes for retrieving, sorting, analysing, predicting, and storing DNA and protein sequence data. Bioinformatics is an evolving discipline, and expert bioinformaticians now use complex software programmes for retrieving, sorting, analysing, predicting, and storing DNA and protein sequence data.

The research of genetic disorders is shifting away from looking at single genes in isolation and toward discovering cellular networks of genes, deciphering their intricate connections, and determining their function in disease. A new era of personally tailored medicine will arise as a result of this. Bioinformatics will guide and assist molecular biologists and clinical researchers in making the most of computational biology’s benefits.

I hope you’ve found some inspiration in my post if you’ve read this far.

Happy Learning….!!!

--

--

## More from MLearning.ai

Data Scientists must think like an artist when finding a solution when creating a piece of code. ⚪️ Artists enjoy working on interesting problems, even if there is no obvious answer ⚪️ linktr.ee/mlearning 🔵 Follow to join our 18K+ Unique DAILY Readers 🟠

## Swetank Pathak

Data Scientist, Sports Physiotherapist and Sports Scientist @ Centre For Sports Science LinkedIn: Swetank Pathak