Tuesday, October 25, 2016

Research Projects

In this section, the research and projects are shared...

(Only principles and ideas are explained in brief since they are IP, please bear with me!)
________________________________________________________________________________

I am basically an Electrical Engineer but my research area is in Biomedical Engineering. My specialized focus is on Imaging, Instrumentation and diagnostic instrumentation; but certainly it is not boundary for my research!

I have worked on
  • Optical imaging for medical diagnosis, 
  • Diabetic foot ulcers, 
  • Photo-acoustic imaging & spectroscopy
  • Microwave imaging 
  • and so on...
Few projects on which I worked previously are briefly explained below.


1) Near infrared imaging and spectroscopy:



In this technique, light in the range of 700-1100 nm wavelength from an electromagnetic spectrum is used as source of energy to shine the biological tissue of interest and reflected or transmitted or backscattered light from the same tissue is detected from the sensors. After the signal conditioning, the output is analysed qualitatively. There are several challenges, governing parameters, selection of suitable instrumentation and understanding the theory of the mechanism in the tissue. All these need to studied and addressed. 

Near Infrared Imaging
Near Infrared Spectroscopy

This method is cost effective and has potential to characterize the tissue; also to detect anomalies present in the tissue such as tumor/haematoma. Therefore, near infrared imaging & spectroscopy is emerging as new diagnostic tool in medical field. 


2) Time correlated single photon counting for optical spectroscopy and imaging applications:
This is based on optical technique for the detection of photons in time domain. The laser light is detected from a photodector and the signal is very weak. Meaning that, only few photons out of input light are able to detect from the tissue. Therefore, a non-linear crystal based on SHG (Second Harmonic generation) is used for counting the few output photons through time correlation analysis. The system is not only cost-effective but also highly sensitive. Its main application is in the detection of breast tumors.
(A*STAR Sponsored Project, NUS, Singapore)


3) Photo-acoustic imaging and spectroscopy:
Optical imaging such as diffuse optical imaging has major drawback of light scattering through the tissue due to which the resolution is poor. When the photons penetrate in to the tissue the light energy is absorbed by the tissue cells. The tissue cells gained energy then propagates in to the surrounding in the form of acoustic waves. So, these acoustic waves are detected from ultrasound sensors from the surface of the tissue. Light energy is used as in input source but the detected output in the form of acoustic waves, hence this technique is called as photo-acoustic method. The input light source may be single wavelength or multi-wavelength laser source emitting in form of short pulses. 

An experimental setup of the PA system is shown in below picture (Courtesy; Pittsburgh University, USA).




4) Microwave imaging system for biomedical applications:
Microwave imaging is a promising technique in which electrical property distribution in the human body are mapped for medical applications. It has been widely researched specifically for early stage detection of breast cancer. Recently, this imaging technique is widely researched because of its high contrast, patient comfort and safety at low cost. An UWB patch antenna transmits the microwave frequency propagating through the target and another the receiving antenna receives the backscatterd microwave signal in the medical band of frequency (2.4 and 5.8 GHz). The signal is measured and analysed through a network analyser. The data is stored in computer and then plotted for electrical parameters to reconstruct image of the object. The patch antenna for transmitter and receiver is designed, simulated and fabricated then used for medical imaging applications. 



5) Medical image segmentation:
Routine clinical examination may not be enough for any conclusive diagnosis of the disease in several occasions. Imaging plays an important role for the clinical diagnostic and therapeutic procedures. The radiographic imaging varies according to their pathologic features. Image segmentation refers to partitioning of an image into distinct regions of interest that contain pixels with similar attributes. The main goal of segmentation is to differentiate the object of interest from the background of image. Thus the segmentation of image may be considered as an important step in image processing techniques. Accurate segmentation of medical images becomes a key step in image processing which can help for suitable planning of diagnosis and treatment of the patient. This technique can guide the radiologists precisely for a specific diagnosis of lesions or delineating pathological regions, thus limiting the differential diagnosis. Several algorithms do the segmentation on images but Watershed method, which is based on natural phenomenon of water flow in lakes, is attracting large number of researchers due to its fast action, simple algorithm to implement in computer and versatile for twin images too. The morphological watershed image segmentation is implemented on MRI images to extract calvarial bone (skullcap) from the scans. It is useful for evaluation of calvarial lesions, calvarial bone grafts and calvarial bone fractured patients that could help in diagnosis and treatments. The same algorithm is extended to segment the cranial vault bone from radiologic images of human head.



6) Early detection of Alzheimer's disease:
Alzheimer’s Disease (AD) is related to human brain which affects a significant fraction of the aged people causing problems with short-term memory, thinking, spatial orientation and behavior and worsening over a period of 20 years ultimately leading to death of the patient. Currently, there is no cure for the disease. Although the onset of AD cannot yet to be stopped or reversed, an early detection and diagnosis could be helpful to the people with AD a better chance to benefit from available drug and non-drug therapies that may improve their health condition and enhance their quality of life. Electroencephalography (EEG) is considered as one of the most important tests for the diagnosis of neurological and neurophysiological disorders which is a measure of the electrical activity of the brain. When the neurons of the brain are activated they produce electrical signal. Neurons are mutually connected in the form of closed neuronal networks. If there is any discontinuity in the path of the network leading to the damage of anatomical and physiological tracking of the circuitry would result in steady decline of cognition that will be an early indication of Alzheimer’s. The study is in progress. The below schematic shows the proposed method of recording the brainwave data.


From this study the probable biomarkers of AD are investigated. The structure of the study is shown in the below figure.


This method is non-invasive, painless, cost-effective and can be repeated several time for thorough investigation to diagnose the disease in its early stage.



7) Diagnosis of Parkinson's disease from SPECT images:
Parkinsonism (PS) is a clinical syndrome described by tremor, bradykinesia, rigidity, and postural instability. The name “parkinsonism” is derived from the progressive neurodegenerative illness called Parkinson’s disease (PD). Although there could be a number of underlying causes, PD remains the most common cause of parkinsonism. The accuracy of the clinical diagnosis of PD is still limited as there is no standard diagnostic test is available. Furthermore the disease is incurable; however a pre-stage diagnosis can help to minimize the progression of the disease. Though there is no specific clinical test to confirm the PD, but imaging of brain is considered as an effective diagnostic procedure since it gives visual insight of the brain in the form of an image. Traditional diagnostic imaging with CT or MRI brain scans does not show changes in the brain but they give anatomical pictures of the brain. To identify the disease in early stages, often functional imaging is preferred than the diagnostic imaging. For the definitive diagnosis of the PD, specialized imaging techniques are required for histologic examination of Intraneuronal Lewy body inclusion of substantia nigra compacta, which are not possible practically in life. However, the studies suggest that the nigrostriatal projection loss is characterized with the PD symptom which is associated with the striated dopamine deficiency. 

Clinically, two technologies are commonly used as nuclear imaging and they are; positron emission tomography (PET) and single photon emission computed tomography (SPECT). They provide the means for detecting in-vivo metabolic and neurochemical changes of PD. These imaging technologies made possible to visualize functional, chemical processes and metabolic activities in the targets by using radiotracers as agents. SPECT imaging is widely accepted clinically to conform the loss of dopamine chemical and this type of scan is known as dopamine transporter scan or DaTSCAN, by making possible to use as an aid that can help diagnosing the PD in early stages. Thorough investigation of the scans could able to detect dopamine transporters. However, for reasonably correct assessment of the images, appropriate algorithms are essential to play the role in the diagnosis of the disease more accurately. In our on-going work, we aim to classify the stage of the PD so that medical expert can know at which stage the PD is present in the patient. In this study, we to use sparse learning techniques to develop some novel classifiers  by utilizing the power of Neural Networks.



8) Human gait evaluation system for rehabilitation of elderly people:
Human gait refers to locomotion achieved through the movement of human limbs. Since the gait is a largely automated motor task it becomes an identity of a person’s style and quality of life. Often, gait can be used as a biometric measure to recognize known persons and classify unknown subjects. Visual inspection of the person’s gait also indicates the health condition of the subject.

Observation of gait is an important aspect of diagnosis that may provide information about several musculoskeletal and neurological conditions. Reliable cognition of gait characteristics over time, continuous monitoring, evaluation and proper analysis have demonstrated their significance in clinical, sports, rehabilitation, personnel training and robotic research. In the field of biomedical engineering, gait analysis and evaluation is attracting the researchers.

Gait is altered by different factors, such as age, gender, working conditions, body weight, etc.  Ageing has great impact on gait and its pattern changes drastically with normal ageing. It is therefore necessary to understand the kinematic and kinetic parameters of gait pattern and hence to take preventive steps against abnormalities related to gait disorders. In this project, it is proposed to observe the gait pattern from normal healthy people and lower limb affected aged people. It is also proposing to suggest suitable corrective measures for improvement in abnormalities. The sketch shows experimental setup to acquire gait pattern data from three directions. 



There are about 25 gait abnormalities. The types of abnormalities while examining the patients are demonstrated in the below video.




9) Mobile phone app for bio-telemetry:
Rapid advancements in communication technology have spread to medicine also. Particularly, smartphone technology has made medical provisioning through mobile systems a reality. Innovations in mobile software application are potential benefits to the public health since the mobile platforms became more user-friendly, computationally powerful and are affordable. The innovative mobile apps can contribute in clinical consultation complementing face-to-face interaction in the health care at lower risk to the public. We have developed and evaluated mobile app for smartphone on Android platform to facilitate interaction between the patient and doctor where the patient seeks advice, diagnosis and treatment from the doctor from remote places. The Graphic User Interface (GUI) display screens of the smartphones are incorporated the medical data needed by the clinician to interpret and respond to information. The scheme of the project and screen-shots from mobile phone are shown below.







The medical application software for quick diagnosis helps the patients to get treatment plans remotely by saving the time of travel to visit the clinic. The doctors can diagnose and treat the patients by this app at their convenient time from their places. Thus the time and cost of both patients and doctors will be saved. The software is user-friendly without involving many operations and no computers or dedicated software are required. It is an open source, freely accessible software to benefit the user whoever need this app. Installation of the app in the smartphone is quite simple and more useful to socio-economically poor people as well as rural dwelling patients.


10) Electrosurgery Unit:
Electrosurgery unit is a helpful tool in all areas of the surgical field from the most basic surgery as wart removal, or hair removal to the most sophisticated surgeries as open heart, orthopedic, and transplant procedures. The main principle of this device is the conversion of electricity into heat.

Also, it is capable of producing a cutting and/or coagulating clinical effect on tissue by the use of alternating current at a high frequency. Voltages and Currents can be changing depending on the desired clinical effect. The idea tried in this project helps to make surgery in short time, decrease the cutting and dangerous coagulation, minimize the loss of blood in surgery and cancellation of the use of sharp surgical tools. The block diagram of the project is shown below.


The unit developed is only prototype and is tested clinically. However, the design features and the product would allow for real tests only standardization following regulatory norms. 

No comments: