Title : Analysis and classification of diseases based on lifestyle with blood group prediction through fingerprint map reading using deep neural network
Abstract:
In the current digital world, hash may consider as a footprint or fingerprint of any digital term but from ancient era human fingerprint considered as the most trustworthy criteria for identification. The fingerprint of a human cannot be change with time even up to death of an individual. Due to the immense potential of fingerprints as an effective method of identification, it opens a lot of possibilities for human science research. This research investigates the problem of blood group identification through fingerprint. The fingerprint can be divided into basic four categories such as Loop, whorl, arch, and composites. Also, the ridge count and different angles in ridgeline are studied by many researchers in their research.
The first proposed method designed using fingerprint minutia feature vector, which prepared using FR core algorithm. It includes the Segmentation, Orientation, Ridge frequency estimation, Binarization, and then finally Thinning processes are applied sequentially; the NIST-4 fingerprint dataset is used to test the feature selection process. The proposed techniques produce better classification performance because there are 21 different minutiae features are extracted in the design of the Multilayer Deep Neural Network which is proposed for the prediction of the blood group of an individual with 90.31% accuracy and only 9.69% misclassification or rejection rate. The model 2 proposed an optimized CNN which is designed as an extension of an AlexNet, that correlates the fingerprint patterns or different features of the fingerprint with the blood group of an individual. The design of proposed CNN used for the prediction of the blood group having noticeable performance with 95.27 % accuracy rate. The diseases based on individual lifestyle which are arises with aging like hypertension, type 2- diabetes and arthritis are analyzed and classified through fingerprint pattern, blood group, age, and different lifestyle habits of an individual. The initial investigation performed on Pima Indian diabetes and hypertension database downloaded from Kaggle. It is further validated using dataset prepared by survey which includes lifestyle eating habits, consumption of alcohol, smoke, exercise routine and work structure etc.
Audience Takeaway:
- The human fingerprint used as digital hash from accent era, and due to its uniqueness as immense potential, fingerprints as an effective method of identification, it opens a lot of possibilities for human science research. This research investigates the problem of blood group identification through fingerprint. The fingerprint cannot only be limited to Loop, whorl, arch, and composites. Also, it has differentridge count and different angles in ridgeline are studied by many researchers in their research, which helps o identify fingerprints more efficiently. One of the main contributions of this work is to extract 21 different minutiae from fingerprint using fingerprint minutiae featuresextraction algorithm and prepared feature vector used to train the ANN which predict blood group of an individual. These algorithms have an accuracy of 90.31%.
- The lifestyle disease is driven by seemingly unrelated causes such as rapid unplannedurbanization, globalization of unhealthy lifestyles and population ageing. Apparentcauses such as raised blood pressure, increased blood glucose, elevated blood lipids andobesity may be representations of deep lying lifestyle habits. In the current era due to busy schedule and unhealthily lifestyle such lifestyle disease arises at any stage of age in individual.
- There are severalrisk factors that lead to the onset and development of such diseases. The various typesof risks can be divided into three primary risk sets: modifiable behavioral risk factors,non-modifiable risk factors and metabolic risk factors, many of which are common forseveral diseases. behavioral risk factors such as excessive use of alcohol, bad foodhabits, eating and smoking tobacco, physical inactivity, wrong body posture anddisturbed biological clock increase the likelihood of lifestyle disease. The modernoccupational setting (desk jobs) and the stress related to work is also being seen as apotent risk factor for lifestyle disease. According to the WHO, more than 7 millionpeople die each year due to the use of tobacco and the fatality rate is projected to increase markedly in the years to come. Excessive use of sodium in the diet causes 4.1 million deaths per year while alcohol intake leads to around 1.65 million deaths due to lifestyle disease. A simple lack of physical activity has been claiming 1.6 million lives annually. To analyze these issues, the proposed research designed a survey which includes age, fingerprint pattern, blood group, BMI index with lifestyle parameters like eating habits, consumption of alcohol, smoke, exercise routine and work structure etc. collect 1024 samples, which are further analyzed to study lifestyle diseases.
- Furthermore, similar studies help to predict diseases at a young age of an individual. Analyzing and classifying communities according to age, blood group, fingerprint patterns, and lifestyle disorders can all be used to assist and prepare for future pandemics, such as COVID-19, in which mankind will be plagued by lifestyle-related diseases like type 2 diabetes and hypertension.
The Relation between Diabetes,Hypertension and Rheumatoid Arthritis Patients and Fingerprint Pattern
- Diabetes Patients
- Surge in arches in diabetes in both genders
- Growth in rate of recurrence of loops and arches and a lessened frequency of whorls especially in mid finger.
- Reduced number of arches in the right hand of male and left hand of female having diabetics, it was more in diabetic males and females than in the controls.
- Growth in radial loop, ulnar loop in both male and female diabetics.
- Increase in frequency of whorls in both types of gender in diabetics.
- Blood pressure/hypertension
- Higher prevalence of whorls and loops are associated with higher level of blood pressure.
- Whorls and loops are prime ridge patterns in hypertensive patients.
- ATD angle showed the mean of angle in patient surge rather than in control group.
- Larger frequency of ridge endings in the thumbs and index fingers.
- Amplified frequency in bifurcations and convergences in the middle, ring, and little fingers.
- Blood pressure/hypertension
- Ulnar loop was the most prominent digital pattern in both genders.
- Decrease in the radial loop in both male and female patients.
- Loops were significantly decreased in the third finger of males and a first and fourth finger of females.
- Decrease in the ulnar loops in both the hands of male and female patients.
- Increase in the whorl pattern in the right hand of male patients and in both the hands of female patients.
- Decrease in the arches of the left hand of female patients.
