Deep Analytics: Technologies for Humanity, AI & Security by Sumit Chakraborty, Suryashis Chakraborty, Kusumita - HTML preview

PLEASE NOTE: This is an HTML preview only and some elements such as links or page numbers may be incorrect.
Download the book in PDF, ePub, Kindle for a complete version.

1. SCOPE

 

Scope Analytics

 

Agents: System analysts, business analysts; scientists, doctors, engineers;

Moves : Critical success factors analysis, Requirements management;

Security parameters: define a set of sustainable development goals on healthcare security, cancercare

Application domains:

    • Cancer : cancer of mind / psycho-oncology, neural control and coordination, chemical coordination and integration, digestive system, respiratory system, cardiovascular system, excretory system, locomotion and movement system, reproductive system;
    • Biomedical technology innovation : artificial pancreas, artificial liver, artificial kidney, artificial limbs, artificial cardiovascular devices, oral insulin;
    • Artificial immune system for epidemic control.

 

Prof. P. Kar is starting the session based on key focus areas such as cancer prevention, proactive approach, reactive approach, bad luck, deep learning, optimal margin classifier, support vector machine, precision medicine, regenerative medicine, integrated medicine, cancer genomics, CNN, intelligent reasoning, bio- medical technology, oral insulin, artificial pancreas, artificial kidney, artificial liver, artificial cardiovascular devices, artificial limbs, laser, surgical robotics, wearable computing, concurrent engineering, technology innovation, epidemic control, pandemic outbreak, Intelligent broadcast, online grievance management system, articial immune mechanism, self-nonself classification, danger signal, clonal selection, hotspot, cluster, social distancing, security intelligence, business intelligence, bio-terrorism, life science supply chain, and healthcare service chain.

This session explores an emerging technology for prediction and prevention of cancer.. The complexity of emerging technology has been analyzed in terms  of scope, system, structure, security, strategy, staff-resources and skill-style-support. Presently, the technology of cancer care is passing through the growth phase of S- Curve. The technology has been analyzed in terms of a set of intelligent strategic moves such as proactive and reactive approach, deep learning, optimal margin classifier, intelligent reasoning and biomedical instrumentation. Intelligent reasoning has been explored in terms of case based reasoning, perception and common sense. It is also essential to adopt a set of effective reactive strategies such as genomics, precision, integrated, regenerative and alternative medicines to fight against cancer. We have presented a deep analytics based cancer prevention mechanism (DACPM) balancing proactive and reactive approaches.  The mechanism defines human biological system from the perspectives of application, computing, networking, data and security schema. This work also analyzes different types of cancer through DACPM.

This session also shows the application of deep analytics ‘7-S’ model on the innovation of bio-medical technology for cancer care. The complexity of biomedical technology has been analyzed in terms of scope, system, structure, security, strategy, staff-resources and skill-style-support. The scope of biomedical technology has been explored in terms of artificial pancreas, artificial liver, artificial kidney, artificial cardiovascular system and artificial limbs. The critical observation is that oral insulin is a rational, simple, practically feasible and safe option as compared to artificial pancreas. It is hard to develop artificial kidney, liver and pancreas which can mimic all the functions of related biological organs. The concept of oral insulin is now at emergence phase of technology life-cycle; artificial cardiovascular devices and limbs. It is rational to adopt proactive and reactive approaches to overcome the constraints of biomedical technology. This work also explores the scope of laser therapy, pervasive and wearable computing and surgical robotics for cancer care. Is it possible to innovate an affordable cheap test kit for cancer care which should be able to detect cancer of human beings by measuring a set of critical health parameters (e.g. blood, urine, sweat, stool, saliva, mucosa, semen etc.) through biosensors in an automated manner? Can AI promote such complex innovations in future?

This session also presents the construction of a deep analytics based cancer prevention mechanism (DACPM) balancing proactive and reactive approaches. It defines human biological system from the perspectives of application, computing, networking, data and security schema of an information system. The strategic moves of DACPM include deep learning, intelligent reasoning, threat analytics, optimal mix of proactive and reactive approaches, rational healthcare payment function and budget plan and adaptive secure multi-party computation. The performance of human biological system is expected to be verified through the properties of adaptive secure multiparty computation: fairness, correctness, accountability, transparency, rationality, trust, commitment; authentication, authorization, correct identification, privacy, audit; safety, reliability, consistency, liveness, deadlock-freeness, reachability, resiliency, robustness and stability of application integration. It analyzes the complexity of the mechanism in terms of computational cost of deep learning algorithm. This work is specifically focused on reasoning nine test cases through DACPM in depth to fight against the epidemic of cancer: The human biological system is assumed to be a computer. It is not a rational thinking that the most of the causes of cancer are due to bad luck; it is still not known enough about the causes and prevention measures of cancer. Deep analytics does not necessarily mean deep learning algorithm, it is also associated with intelligent reasoning – analytical, logical, common sense, case based reasoning and also perception to fight against the epidemic of cancer. A human agent must have common sense healthcare knowledge base for proper biological system control through intelligent self-assessment, self-confidence, life-style, diet control and right decision making at right time. It demands the necessity of learning the basic concept of reasoning and common sense healthcare through an effective knowledge management system based on deep analytics and intelligent broadcast communication.

The basic objective of this session is to generate a rational cancer prevention plan subject to financial constraints. The work has reviewed the relevant literature on cancer, oncology and deep learning and has adopted analogical reasoning as research methodology.

Technological innovations are practical implementation of creative novel ideas into new biomedical devices. Many potential ideas pass through the wide end of an innovation funnel but very few may become successful, profitable, economically and technically feasible products in future. It is an interesting research agenda whether deep analytics may be an effective tool for the diffusion of biomedical technology in future. It is a multi-dimensional analysis wherein seven factors must be integrated, coordinated and synchronized. This session shows the application of deep analytics ‘7-S’ model on innovation of bio-medical technology.

Recently, there is a trend of cross fertilization between five disciplines: medical science, management information system, artificial intelligence, artificial neural network and management science. This work is associated with the problem of cancer prevention. Cancer is a costly, global and complex problem; it results a major obstacle to human development and well-being. The attack of cancer has increased from 12.7 million (2008) to 14.1 million (2012) and this trend is projected to continue about 25 million cases over next two decades; the greatest impact will  be in low and middle income ill equipped countries. The future of a cancer patient depends on the living zone. In less economically developed countries, cancer is diagnosed at more advanced stages while access to effective treatment is limited or unavailable. The highest-income countries often struggle with the spiraling costs of cancer treatment and care. Cancer has a social cost, human potential is lost and cancer care has an escalating economic impact. It is essential to identify the causes and prevention strategies for cancer control.

Let us first look at bio-statistics of cancer. It is a major cause of morbidity and mortality, with about 14 million new cases and 8 million deaths in 2012, affecting populations in all countries and all regions. Among men, five most common sites of cancer were lung (16.7%), prostate (15.0%), colorectum (10.0%), stomach (8.5%), and liver (7.5%). Among women, five most common sites of cancer were breast (25.2%), colorectum (9.2%), lung (8.7%), cervix (7.9%), and stomach (4.8%). There were 8.7 million people (older than 15 years) alive with cancer diagnosed in the previous year, 22.0 million in the previous 3 years, and 32.6 million in previous 5 years. The worldwide estimate for the number of cancers diagnosed in childhood (ages 0–14 years) in 2012 is 165000 (95000 in boys and 70000 in girls). The highest incidence rates are associated with high income countries of North America and western Europe, Japan, Korea, Australia, and New Zealand. More than 60% of cases and 70% of deaths occur in Africa, Asia, and Central and South America. Cancers are caused by mutations that may be inherited or caused by environmental factors or DNA replication errors.

Dr. Kar has defined the structure of human biological system from the perspectives of application, computing, networking, data and security schema of an information system. The application schema is related to the function and features of a specific biological system. The networking schema is related to the configuration of the system such as nodes and interconnections among the nodes. The computing schema deals with the protocol, process, procedure and mechanisms of a system and its various components. The data schema is associated with various entities, their attributes and interrelationships, inputs and output of a system. The security schema verifies the disorders of the system and protects the system through various means such as vaccination precision, regenerative and integrated medicine, chemotherapy and laser.

In this session, the scope of cancer has been explored in terms of (i) cancer of mind, (ii) neural control and coordination : brain cancer, (iii) chemical coordination and integration : breast cancer (iv) digestion and absorption : liver, pancreas, stomach and colorectal cancer (v) respiratory : lung cancer, (vi) body fluids circulation : blood cancer, (vii) excretory : renal cancer and urinary bladder cancer, (viii) locomotion and movement: bone cancer and (ix) reproductive system : ovarian and testis cancer.

Dr, Rina Brown and Prof. Bob Taylor have explored the scope of emerging biomedical technologies in terms of artificial pancreas vs. oral insulin, artificial kidney, artificial liver, artificial cardiovascular devices (e.g. heart valves, stents, pace makers, artficial limbs, arms, legs and laser therapy. The scope of innovation on biomedical technology should be explored rationally through intelligent analysis of the basic objectives, goals, needs, constraints and mechanisms; strength, weakness, opportunities and threats of various strategic options. First, it is essential to understand the mechanisms of various human organs (e.g. pancreas, liver, kidney, heart, limb, brain) in terms of human physiology, input, output, feedback control, function, process, chemical reaction, secretion of enzymes and hormones, coordination, integration and system performance. Next, it is rational to analyze whether it is practically feasible to make artificial organs which can mimic various functions of biological organs of human body in the best possible ways. The scope of biomedical technology spans over several domains such as artificial pancreas, liver, kidney, cardiovascular system and limbs

Artificial pancreas: Let us first do scope analysis on oral insulin vs. artificial pancreas for the treatment of diabetes; which is more feasible technology innovation and why? Pancreas synthesizes insulin which extracts glucose from carbohydrate for the supply of energy and storage. It controls blood sugar level and prevents hyperglycemia or hypoglycemia. Insulin is a collection of 51 amino acids with two chains A (21 amino acid) and chain B (30 amino acid) linked by disulfide bridges. Diabetes is a chronic disease, it arises when sufficient amount of insulin is not produced by the pancreas (Type 1 diabetes) or insulin which is formed is  not utilized properly by the body (Type 2 diabetes). It leads to an elevation of blood glucose level (hyperglycemia). Diabetes is the most common endocrine disorder. It is a real challenge to find out effective administration and delivery mechanism of Insulin. Subcutaneous (SC) route may lead to hyperinsulinemia. Repeated injections if insulin may result various types of health problems such as lipoatrophy or lipohypertrophy, peripheral hyperinsulinemia, peripheral hypertension, atherosclerosis, cancer, hypoglycaemia and other adverse metabolic effects. Can an artificial pancreas mimic all the functions of a biological pancreas? It is essential to explore alternative route such as oral insulin which mimics the typical insulin pathway within the body after endogenous secretion subject to various constraint such as good bowel absorption and very low oral bioavailability of insulin.

Artificial liver : Next, let us consider the scope analysis of artificial liver. Liver is a complex organ doing various vital functions such as synthesis, detoxification and regulation; its failure may result a life threatening condition. Liver failure (LF) can either occur as acute liver failure (ALF) due to intoxication or as acute-on-chronic liver failure (AoCLF). The common symptoms are icterus, hepatic encephalopathy and impairment of coagulation and may result even multi organ failure. In case of liver failure, water-soluble toxins (e.g. ammonia) and albumin-bound toxins (e.g. bilirubin, amino and fatty acids) may accumulate and cause encephalopathy and dysfunction of other organs. Detoxification and regulation can be addressed by artificial devices similar to dialysis, the synthetic function of the liver can only be provided by living cells. We have done analysis on strength, weakness, threats and opportunities of artificial liver and have also outlined a liver protection mechanism by adopting an optimal mix of proactive and reactive approaches.

Artificial kidney: Next, let us consider the innovation of artificial kidney. There are various therapies of kidney problems [e.g. end-stage renal disease (ESRD, continuous renal-replacement therapy (CRRT)] which cause sepsis, systemic inflammatory response syndrome, acute respiratory distress syndrome, congestive heart failure, tumorlysis syndrome and genetic metabolic disturbances. The dominant therapies are hemodialysis and hemofiltration. An artificial kidney should perform three physical processes efficiently that determine the removal rate for uremic toxins through membrane-based devices: convection removes toxin through a semipermeable membrane; diffusion removes smaller molecules with high diffusion coefficients and adsorption.

Artificial cardiovascular devices: Next, let us consider the technological innovation of artificial cardiovascular devices. Cardiovascular disease (CVD) is the leading cause of death worldwide. In this domain, the technological innovation is facing several challenges such as improved device function (e.g. cardiac valves, stents, pacemakers and defibrillators, vascular grafts, hemodialyzers, catheters, circulatory support devices and blood oxygenators), complex and challenging cardiovascular surgical procedures (e.g. open- heart surgery), medical therapies (e.g. dialysis), valve replacement problems (e.g. thromboembolism, hemolysis, paravalvular regurgitation, endocarditis and structural failure of the valve); artificial heart valves design (e.g. percutaneous or minimally invasive valve implantation and tissue engineering), progress in stent technology to reduce restenosis and improvement of stent outcome, development of pacemakers, cardioverter-defibrillator (AICD), cardiac electrophysiologic devices from the perspectives of improved device function, dual chamber activity, advances in technology and implantation techniques, development of artificial lung for acute cardiopulmonary bypass and improved biocompatibility.

Artificial limbs: The basic objective of prosthetics research is to design and develop artificial arms, hands and legs which can be used flexibly with physiological speeds- of response and strength and controlled almost without thought. The current state  is basically a tool rather than a true limb replacement. The prosthesis is an interchangeable device that is worn and used as needed and then ignored. The major constraints of prostheses are weight, power, size and sufficient number of appropriate control sources to control the requisite number of degrees of freedom. The system requires better sensors, actuators and multifunctional control mechanisms. The basic building blocks of artificial limbs are mechatronics and robotics; current prosthetic components and interface techniques are still a long way from realizing the aforesaid objectives.

 

Dr. Muller is exploring the scope of various types of epidemic and pandemic outbreak and is suggesting a set of intelligent strategic moves as countermeasures. It is rational to adopt an efficient system; an optimal mix of e-governance (e.g. online grievance management system), broadcast communication protocol and artificial immune mechanism to fight against natural disaster, epidemic and pandemic outbreak. It is essential to analyzes security, strategy, staff-resources and skill-style- support for efficient coordination and collaboration in corporate governance. There is threat of bio-terrorism on the soft targets (e.g. life-science supply chain and healthcare service chain). Is the conflict between security intelligence and business intelligence inevitable?

Epidemic and pandemic is a critical causal factor of poverty of human society globally. Can emerging technologies be used to fight against such calamities effectively? Dr. Muller is xploring the scope of epidemic and pandemic outbreak in depth; Pandemic is more dangerous than epidemic. When an epidemic spreads globally, it is called pandemic. In case of epidemic, a disease spreads at very fast rate witin a particular period among one or more communities. For example, WHO has recently declared the outbreak of novel Corona virus as Pandemic, but it is controllable. There are other various types of threats of epidemic globally due to environmental pollution such as air, water, soil, light and sound pollution. The issues have been already discussed during session 2.

Epidemic may happen due to air pollution like dust at construction sites and industrial plants; smoke from vehicles; paste control problem (e.g. mosquitoes, flies), malnutrition in slum areas, improper cleaning of garbages and stool of street animals? Epidemic may occur due to water pollution in supply of dirty drinking tap water caused by leakage in pipelines, contamination and jerms in water storage system, malfunctioning of tube wells, water filtering problem, mixing of water from drainage system and tap water pipeline, unprotected selling of unhealthy food (e.g. oily spicy biriyani) and beverages at retail outlets and by hawkers being contaminated by flies; risks of diahorrea, stomach upset and loose motion. Epidemic due to soil pollution and earthquake caused by random digging of soil for construction projects, erosion of soil at riverbeds; jamming in drains due to plastics, improper cleaning of drainage and sewage system; There is threat of epidemic due to light pollution in slum areas, unplanned urban development planning, blockage of sufficient sunlight into residential areas (e.g. houses, flats, multi-storied buildings) Epidemic due to sound pollution may be caused by playing loud and wild music, fireworks, activities at construction sites and industrial belts.