Intelligent Medical Technologies and Biomedical Engineering: Tools and Applications

This book is basically the result of 20 years of teaching the material to Electronics communication, Information Technology, Computer Science Engineering, Bioinformatics, Biotechnology Engineering and biomedical engineering students. Some of the "students" have also been faculty and post graduate. I am very grateful to them for their patience and tolerance as it progressed from crude notes to its present form.

We describe in chapter 1, Digoxin (DGX) is drug used to control signs and symptoms involved in congestive heart failure and atrial fibrillation. Due to its narrow therapeutic range more than 10% of the patients treated with DGX can suffer toxic effects, but it is estimated that half of the cases of digitalis toxicity could be prevented. Software for identifying patients at risk in order to implement preventive measures to avoid signs and symptoms of digitalis toxicity was developed.

We describe in chapter 2, Natural Orifice Transluminal Endoscopic Surgery (NOTES) has advantages in reducing postoperative abdominal wall pain, wound infection, hernia formation and adhesions. In this chapter, we proposed a Master and Slave Transluminal Endoscopic Robot (MASTER) with 9 degree of freedom (DOF) end effectors which are long and flexible so as to enhance endoscopic procedures and NOTES. The robotic system consists of a master console, microprocessor, actuator housing and slave manipulators.

We describe in chapter 3, the goal of this chapter is to introduce the electrospinning technique - a materials processing technique that uses an electric field to draw a polymer solution into ultra-fine fibers. Further, this chapter aims to provide pertinent information to researchers who intend to use electrospinning in their research. The electrospinning technique was invented a few decades ago and has recently been adapted for biomedical applications. Electrospinning is now widely used in the biomedical field due in large part to its capability of fabricating ultra-fine fibers with unique physical and biological properties.

We describe in chapter 4, Breast cancer is the second leading cause of cancer deaths in women and occurring one out of eight women in worldwide. Early detection is an effective way to diagnose and manage breast cancer. Currently there are three methods used in breast cancer treatment: mammography, FNA (fine needle aspirate) and surgical biopsy. In this paper we found that three techniques are mostly seen in the literature of breast cancer. Some of the specific existing approaches are specifically described for breast cancer and the obtained results are evaluated and noted carefully in order to give a clear view of the topic to the reader.

We describe in chapter 5, Mastoidectomy is a core surgical procedure in otologic surgery. It is believed that the procedure is performed by different surgeons with some variations. Also, all surgeons use a finite number of fundamental surgical actions to complete the procedure. Here, we sought to identify the fundamental surgical actions (called Action Primitives, APs) constituting mastoidectomy and determine transition boundaries between those APs. Our motivation for this work is both to delineate the APs necessary to complete a mastoidectomy and to optimize and potentially automate major components of the surgical process. Here we present a novel approach to developing methods for parsing raw data (position and orientation of the surgical tool and end-effector force) into a sequence of surgical actions (APs) that can be used by a robot in the future. In this chapter we present results from our initial investigation on detecting transition boundaries and identifying APs involved in mastoidectomy.

We describe in chapter 6, biological signal has the specialty over the other physical signals in terms of the remoteness of their source. As the analysis of biological signal plays an important role in medical decision making, the subject information should be accurate and reliable. Though these types of signals are better studied by trained and experienced medical practitioners, long term continuous monitoring and automatic decision making is better done by a computerized expert system. So it becomes a part of engineering under automatic signal processing and analysis studies. ECG being the most vital physiological signal, its acquisition technique, noise and artifacts elimination methodologies are discussed in this chapter. A brief description on automatic classification techniques is also given.

We describe in chapter 7, this work proposes the use of two neural models for data analysis of hemodialysis patients with end-stage renal disease. There are two main goals: firstly, the knowledge extraction from a database using Self-Organizing Maps (SOMs); and secondly, to provide an accurate prediction of Hb levels next month. The achieved results show the ability of SOM to profile different behaviors present in the database. Regarding the prediction task, the obtained neural models are equivalent (p<0.05) to the linear ones in the case of EPO beta but they differ in the case of Darbepoietin. In terms of the committed error, the prediction is relatively accurate; Mean Absolute Errors are lower than 0.5 g/dl for both kinds of EPO, thus ensuring that the obtained prediction models can be used to get the goal of maintaining patients stable, within a target range of Hemoglobin (Hb) (usually, between 10 and 12 g/dl). The accurate prediction provides sufficient knowledge to be even more demanding in the target range, being able to work with narrower, more demanding ranges; e.g., 10-11 g/dl for female patients and 11-12 g/dl for male patients.

We describe in chapter 8, Venom is toxic substance developed in an animal. The basic purpose of venom is to defend self by immobilizing the opponent, be it a predator or a prey. The nature has blessed many animals with venom. Snakes and spiders house venom in their mouth and scorpions keep it in their tail whereas, honeybees, wasps and other insects have developed special stingers for venom delivery. Venom is primarily not meant against human beings. But, in a survival conflict, one may interact with venom. The innate immune system and the delicate feedback mechanism respond to venom and manifest in the form of various biosignals. ECG being one of the most susceptible and quick signal responds to venom interactions and the responses can be identified as general as well as species specific. Various ECG manifestations of common venom interactions and Digital Signal Processing techniques for ECG analysis are presented here with the objective of improvement in the medical management of victims of venom interactions.

We describe in chapter 9, the applications of Hybrid Soft Computing approaches that seem to have completely replaced the traditional uni-system approaches. The added abilities that come from the hybrid approaches motivate their use in every system. We find various new approaches being applied to the field of Bio-Medical Engineering as well as many new models being proposed. At this juncture, we study the effectiveness of various new hybrid approaches in the field of Bio-medicals. PIMA Indian diabetes database has been used for this purpose from the UCI Machine Learning Repository. The basic aim is to compare the various hybrid approaches from the recent literature and compare their performances. We study 3 major Hybrid Systems and standard Back Propagation Algorithm for this purpose. These are Adaptive Neuro Fuzzy Inference Systems, Ensembles and Evolutionary Artificial Neural Networks. We also try to explain the results from our theoretical understanding of the individual Hybrid Systems.

We describe in chapter 10, are persistently getting diversified, evolved and thus require rapid vaccine development methodologies. Complex archaic methods require painstaking efforts over a larger time span. In silico methods are utilized these days to screen potent antigens. This approach paves a future way to reduce the number of wet-lab experiments.  Its importance is also highlighted by the rapid scattering of diseases and evolution of their variants. This cited work explains immunological aspects and algorithms applied for epitope selection, with their practical problems. Current techniques used to model the native state conformation of epitopes are then described with their fundamental problems. So, the algorithms to validate such models are discussed. Techniques to screen effective potent drugs against the target epitopes are then considered. Lastly, scope for further research in developing better methodologies is highlighted.

We describe in chapter 11, the application of an intelligent machine learning technique (Support Vector Machines, SVM) to diagnose the patients with sleep apnea syndrome using Electrocardiogram (ECG) signal. Sleep apnea syndrome is a medical condition caused by sleep apnoea which is defined as the cessation of breathing for short periods during sleep. First, the importance of early diagnosis and treatment of sleep apnea syndrome are presented. This is followed by an introduction to the design of a home diagnostic model for predicating sleep apnoea syndrome from electrocardiogram recordings.

We describe in chapter 12, the science of robotics began as a form of entertainment and has evolved into a technology that used in various fields of computers, automotives, entertainment, ocean/space exploration, and medicine. In medicine and more specifically surgery, robotic technology has become a valuable surgical alternative to provide the new methods of surgery with the advantages of traditional open surgery approaches. In this chapter we discuss minimally invasive surgery or MIS and robotic surgery and introduce different robotic surgery systems and technologies in the world.

We describe in chapter 13, explain a new method of generating a discrete version of the continuous Hermite functions to biomedical signals and images that are digital. The process is computationally simple, and provides a set of vectors that share many of the properties of the continuous Hermite functions that are otherwise lost in the discretization process.   Specific applications of this Discrete Hermite Transform (DHmT) are provided that include (i) monitoring ischemia from electrocardiograms, (ii) artifact removal from electroencephalogram, and (iii) ultra wide band wireless monitoring of respiration rate.   This DHmT holds much promise in biomedical signal and image processing.

We describe in chapter 14, the main goal of the chapter is to emphasize, by the enumeration of some important results, activities, systems, groups, institutions, theories, the benefits ontologies bring to the e-Health and biomedicine domains and the effort already given in this respect. The second goal of the chapter is to describe (by means of a case study) the representation and integration of the semantic and modeling layers of a system based on ontologies for the prevention of occupational risks. This system is seen as an application of ontologies in the occupational health domain.