Patil Vijaykumar, Ingle D R
Bharati Vidyapeeth College of Engineering, University of Mumbai, Navi Mumbai, India.
Artif Intell Rev. 2021;54(3):1803-1839. doi: 10.1007/s10462-020-09891-w. Epub 2020 Aug 18.
In the current era of the digital world, the hash of any digital means considered as a footprint or fingerprint of any digital term but from the ancient era, human fingerprint considered as the most trustworthy criteria for identification and it also cannot be changed with time even up to the death of an individual. In the court of law, fingerprint-proof is undeniably the most dependable and acceptable evidence to date. Fingerprint designs are exclusive in each human and the chance of two individuals having identical fingerprints is an exceptional case about one in sixty-four thousand million also the fingerprint minutiae patterns of the undistinguishable twins are different, and the ridge pattern of each fingertip remain unchanged from birth to till death. Fingerprints can be divided into basic four categories i.e. Loop, whorl, arch, and composites, nevertheless, there are more than 100 interleaved ridge and valleys physiognomies, called Galton's details, in a single rolled fingerprint. Due to the immense potential of fingerprints as an effective method of identification, the present research paper tries to investigate the problem of blood group identification and analysis of diseases those arises with aging like hypertension, type 2-diabetes and arthritis from a fingerprint by analyzing their patterns correlation with blood group and age of an individual. The work has been driven by studies of anthropometry, biometric trademark, and pattern recognition proposing that it is possible to predict blood group using fingerprint map reading. Dermatoglyphics as a diagnostic aid used from ancient eras and now it is well established in number of diseases which have strong hereditary basis and is employed as a method for screening for abnormal anomalies. Apart from its use in predicting the diagnosis of disease; dermatoglyphics is also used in forensic medicine in individual identification, physical anthropology, human genetics and medicine. However, the Machine and Deep Learning techniques, if used for fingerprint minutiae patterns to be trained by Neural Network for blood group prediction and classification of common clinical diseases arises with aging based on lifestyle would be an unusual research work.
在当今数字世界的时代，任何数字手段的哈希值都被视为任何数字术语的足迹或指纹，但从古代起，人类指纹就被视为最可靠的身份识别标准，而且即使到一个人去世，它也不会随时间而改变。在法庭上，指纹证据无疑是迄今为止最可靠且可接受的证据。每个人的指纹图案都是独一无二的，两个人拥有相同指纹的概率是极其罕见的，约为六十四亿分之一，即使是难以区分的双胞胎，其指纹细节图案也不同，而且每个指尖的纹路图案从出生到死亡都保持不变。指纹可大致分为四类，即箕形纹、斗形纹、弓形纹和复合纹，然而，在一枚完整的指纹中，有超过100种交错的嵴和谷的形态，即所谓的高尔顿细节。由于指纹作为一种有效的身份识别方法具有巨大潜力，本研究论文试图通过分析指纹图案与个人血型和年龄的相关性，来研究从指纹识别血型以及分析与衰老相关疾病（如高血压、2型糖尿病和关节炎）的问题。这项工作受到人体测量学、生物特征识别和模式识别研究的推动，表明通过读取指纹图谱来预测血型是有可能的。皮纹学作为一种诊断辅助手段，从古代就开始使用，如今在许多具有强大遗传基础的疾病中已得到充分确立，并被用作筛查异常情况的一种方法。除了用于预测疾病诊断外，皮纹学还用于法医学中的个人识别、体质人类学、人类遗传学和医学领域。然而，将机器学习和深度学习技术用于通过神经网络训练指纹细节图案，以预测血型并基于生活方式对与衰老相关的常见临床疾病进行分类，将是一项不同寻常的研究工作。
