Showing papers presented at "Biomedical Engineering International Conference in 2016"

Smart wheelchair based on eye tracking

Abstract: This project is a smart wheelchair based on eye tracking which is designed for people with locomotor disabilities. The add-on controlled module can be used with any electrical wheelchair. The smart wheel chair consists of four modules including imaging processing module, wheelchair-controlled module, SMS manager module and appliance-controlled module. The image processing module comprises of a webcam installed on the eyeglass and C++ customized image processing software. The captured image which is transmitted to raspberry Pi microcontroller will be processed using OpenCV to derive the 2D direction of eye ball. The coordinate of eyeball movement is then wirelessly transmitted to wheelchair-controlled module to control the movement of wheel chair. The wheelchair-controlled module is two dimensional rotating stages that installed to the joystick of the electrical wheelchair to replace the manual control of the wheelchair. The motion of eyeball is also used as the cursor control on the raspberry Pi screen to control the operation of some equipped appliance and send message to smart phone.

Detection of lesions and classification of diabetic retinopathy using fundus images

Abstract: Diabetes retinopathy is a retinal disease that is affected by diabetes on the eyes. The main risk of the disease can lead to blindness. Detection the disease at early stage can rescue the patients from loss of vision. The major purpose of this paper is to automatically detect as well as to classify the severity of diabetic retinopathy. At first, the lesions on the retina especially blood vessels, exudates and microaneurysms are extracted. Features such as area, perimeter and count from these lesions are used to classify the stages of the disease by applying artificial neural network (ANN). We used 214 fundus images from DIARECTDB1 and local databases. We found that the system can give the classification accuracy of 96% and it supports a great help to ophthalmologists.

Teeth segmentation from dental x-ray image by template matching

Abstract: 3D digital dental model is used for orthodontic treatment. Teeth segmentation is an important process for a 3D model construction. In this paper, we proposes a method to segment a teeth from a panoramic dental x-ray image. The proposed method consists of three steps: tooth area identification, template matching, and teeth area segmentation. First, to identify tooth area, the Otsu's threshold and Mahalanobis distance technique are used. Second, teeth template images with different image size are used to match with a teeth in x-ray image. Finally, overlap area from matching multiple templates is used to segment a teeth. To test performance of the proposed method, twenty-five dental images are used. Totally, there are 450 single-rooted tooth, and 250 double-rooted tooth. An accuracy are 42.20 and 49.04 percent for single-root and double-root teeth, respectively. An accuracy should be improved in further. However, this is first fully automatic teeth segment method by using panoramic dental x-ray image.

The effect of average filter for complementary filter and Kalman filter based on measurement angle

Abstract: The objective of this research project is to construct an Inertial Measurement Units (IMU), which will design filter for users use equation on average filter complementary filter and Kalman filter and use algorithm on MCU. The sensor design, we use low-cost sensor on gyroscope and accelerometer, to measurement angle of system. The filter structures of the system, based on complementary filter and Kalman filter and use series average filter for measurement. This project will get data form experimental results to develop on simulation to illustrate the effectiveness of each filter scheme. Simulation study of average filter on different parameter to using on complementary filter and Kalman filter.

Social data shoes for gait monitoring of elderly people in smart home

Abstract: Gait monitoring technology has recently become of increasing interest in the biometric as well as biomedical fields for observation of the human movement, especially walking that can refer to physical status of individuals. For this objective, data shoes own many advantages such as cheaper cost and rich of direct information from the walking. In this paper, we have developed a data shoe system for gait monitoring in home area. To observe gait behaviors, the sensor suite includes five force sensitive resistors (FSRs) which were installed on the insole of the shoe. Zigbee wireless communication technology was used as low-cost data transfer between the sensor suite and the receiver system which is USB-connected to a computer. The summary of the gait data can be submitted and displayed on social media such as Facebook in order that relatives or care-takers can monitor the wearer closely. Principal component analysis (PCA) pattern recognition of the experimental data has shown that this system can classify normal and abnormal walking patterns in a group of elderly volunteers. The integration of sensors, wireless technology and social ability with computer software could make the social data shoe system monitor the gait behaviors during the wearing time.

Self-screening for diabetes by sniffing urine samples based on a hand-held electronic nose

Abstract: Biomedical equipment for early health screening and classification of diseases are useful to assist in the medical diagnosis and maintaining health. Diabetes is a common and increasing disease worldwide. Thus, the development of a new screening method using a urine odor detection device for identifying type 2 diabetes mellitus should be very beneficial. In this study, we report a self-monitoring system to detect specific sweet-smelling urine odor made from four polymer/functionalized-SWCNTs nanocomposites gas sensors, also known as a hand-held e-nose device. The sensitivity and specificity of gas sensing units were characterized and evaluated in the static chamber contains nitrogen gas at room temperature (25°C). Six volatile organic compounds (VOCs) such as ammonia, ethyl methyl ketone, butyric acid, acetic acid, acetone and water were used as biomarkers to represent the many types of urinary volatile compounds found in diabetes mellitus and were used to examine the performance of our sensors. Preliminary evaluation of urine odor sensors with a hand held e-nose device showed that these sensors have high response to ammonia, ethyl methyl ketone and acetone, respectively. Furthermore, the hand-held e-nose was able to discriminate between urinary odors from four diabetic patients and three healthy volunteers. The individual's specific urine odor (urine's odor print) from seven volunteers was confirmed by cluster analysis (CA) method and principal component analysis (PCA) which successfully classified 99.5%. Therefore, this personal diagnostic screening device is likely to be useful for real-time self-monitoring of urine odor in patients with diabetes mellitus and those who are at high risk of developing diabetes disease. In addition, this screening method is painlessness, non-invasive, safe and convenient for integrating health tracking into the future smart home.

Screening and discrimination of Hepatocellular carcinoma patients by testing exhaled breath with smart devices using composite polymer/carbon nanotube gas sensors

Abstract: There is currently great interest in the medical application of electronic nose and chemical sensors, especially in the area of early diagnosis and screening of diseases. In this study, a pocket electronic nose based on eight nanocomposite gas sensors made of polymer and functionalized single-walled carbon nanotubes (f-SWCNTs) was shown to be capable to discriminate between the two sample groups of hepatocellular carcinoma (HCC) patients and healthy control subjects. Polymer/f-SWCNTs sensor-integrated electronic nose system has been designed and fabricated to be suitable for exhaled breath detection. This chemical gas sensor array has a good sensitivity to a broad range of volatile organic compounds (VOCs), sufficiently to cover the chemical species contained in the human exhaled breath such as acetone, ammonia, methyl-ethyl-ketone, and toluene (excluding water that has negligible impact on sensitivity of the sensors). The obtained results demonstrate that the e-nose has a potential to discriminate the patterns of exhaled breath odor from five healthy controls from five HCC patients, as analyzed by the principal component analysis (PCA) with 95% of the confidence level. In the near future, this approach may become very useful in clinical application to serve as a non-invasive device for screening patients with early-stage liver cancer.

Thai sign language recognition by using geometric invariant feature and ANN classification

Abstract: Hand sign language is the primary communication tool for people with hearing-impaired or deaf. People can use it to communicate effectively but the challenge is to communicate with the computer. Human computer interaction (HCI) will have a positive impact on their use. Thus, this is to bring the hand gestures in HCI as an important research area. This research focuses on 2D image recognition utilizing an evolved geometric invariant feature and also have developed a two-layer feedforward neural network to identify and translate hand gesture pose of the 42 letters in the Thai Sign Language (TSL) alphabet to Thai alphabets. We designed glove with six different colored markers for using in the experiment. The result shows that this system is able to recognize 42 TSL alphabets with an average accuracy of 96.19 %.

Stress and office-syndrome detection using EEG, HRV and hand movement

Abstract: Stress and Office syndrome are a serious problem that affects a large number of people and tend to develop health problems that can interfere with work and quality of life. In fact, the Institute reports that up to 90 percent clinical studies have shown that stress and office syndrome are a major cause of cardiovascular disease, depression, suicide and substance abuse. Treatment's cost are more than ten million baht each year in healthcare expenses. According to these problems we designed to use wearable technology which can be tracked and managed to helping them reduce their problems. Low-cost single dry-sensor EEG Neurosky headset and intelligent watch made by Arduino was used in this project. The EEG signal, hear rate variation and hand movement are analyzed to indicate stress level and the Office syndrome can be detected by intelligent watch. The goals of this research is to detect stress and Office syndrome and reduce chance to be the diseases.

Conformational transitions of thai silk fibroin secondary structures

Abstract: Secondary structures of silk fibroin (SF) related directly to their gelling, degradability, mechanical properties which affects its applications. The current works investigated the changes of Thai SF's secondary structure during the process of film fabrication. Thai SF from domestic silk cocoons, Bombyx mori (Nangnoi Sisaket 1) were dissolved into water soluble protein using two solvent systems, 9.3 M LiBr solution, and Ajisawa's reagent (CaCl 2 :water:EtOH at 1: 8: 2 by mole). The SF solution was subjected to liquid nitrogen followed by lyophilization to preserve the protein's structure. Films were casted from the SF solutions and were immersed in ethanol to regenerate the protein's original conformations. Attenuated totally reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) was used to quantify relative contents of protein's secondary structures altered by each step of the processes. Degummed silk fibers had the highest beta sheet crystallinity of 59.1%. Random coils and beta sheet structures of the SF prepared with Ajisawa's reagent were at 41.5% and 18.2%, and those prepared with LiBr solution were at 45.8% and 14.4%, respectively. The non-treated SF films consisted of 33.8–35.0% and 23.2–25.9% of beta sheets and random coils, respectively, despite of their preparation processes. Regeneration of the SF using ethanol immersion increased the beta sheet contents by 10–12% compared to the non-treated ones. Relative contents of beta turn, alpha helix, and tyrosine side chains were unchanged. The EtOH-regenerated Thai SF films contained 11–17% lower in beta sheet structures than the native fiber, suggesting the silk II formation was partially regained.

Design and development of a smart continuous passive motion device for knee rehabilitation

Abstract: Continuous Passive Motion (CPM) device is widely used for knee rehabilitation to recover the range of motion or to lessen edema and swelling of the knee following injuries or surgeries. The objective of this work was to design and develop a smart continuous passive motion (SCPM) for knee rehabilitation after the total knee arthroplasty (TKA). A microcontroller with built-in 7-inch touch-screen LCD was used to control the developed device. There are 4 main operating modes, which are Manual, Auto Run, Intermittent and Progressive modes. In addition, the developed device is capable of flexing/extending the knee joint in the range of motion from 0 to 120 degrees. A rotary encoder was installed to ensure the accuracy of the knee joint angle. In addition, patient data and operating data can also be stored on a memory card. To verify the performance of the developed SCPM, the degrees of motion were tested by comparing the movement angles of the developed device with the angles measured from goniometer. The results show that the errors were found to be within 1%. The developed SCPM was also evaluated by physiotherapists. The results show that it gives better benefits associated with its conventional counterparts since it provides more user friendly interface for the controller and it also has better operating modes (Intermittent and Progressive modes) for the safety of the patients.

Foot plantar pressure measurement system using optical sensor

Abstract: Foot pressure measurement is necessary for classifying disorders of the foot and designing insole for individual person. This present work uses optical sensor (webcam) to capture the foot-pressure image. Image processing on Raspberry Pi with OpenCV library is applied to process image and color coding corresponding to foot pressure. The hardware system uses the transparent acrylic plate and uses the steel as a base of the acrylic plate. The glossy white paper is placed on the top of the transparent acrylic plate covering with polypropylene sheet on the system to block light from outside. Light in the system is released from LED strip entering from a side of the acrylic plate. The scattered light occurred in acrylic plate from the foot pressing were recorded by the webcams. The four webcams placed below facing upward for collecting images (2 cameras for each foot) and sending to Raspberry Pi. Raspberry Pi will perform image process including image enhancement and image stitching and image displaying. The result can be used to classifying foot type and find methods to prevent the occurrence of for disorders.

Initial investigation of channel capacity for galvanic coupling human body communication

Abstract: Human Body Communication (HBC), which utilizes the human body as communication channel to convey health informatics, is a prospective communication technique for implantable and surface-mounted medical devices. In this article, we have applied Shannon's theorem to derive the channel capacity for the general galvanic coupling HBC channel. The channel characteristics and channel model of general galvanic coupling HBC channel are firstly discussed. Based on this channel model, the channel capacity is derived. By using the water-filling algorithm, the HBC channel capacity is estimated and the suitable frequency bands for low power data transmission in both on-body and in-body HBC channels are found. The results can provide suggestion for better system design.

Development of a cost-effective ecg monitor for cardiac arrhythmia detection using heart rate variability

Abstract: Ischemic Heart Disease (IHD) and stroke are statistically the leading causes of death world-wide. Both diseases deal with various types of cardiac arrhythmias, e.g. premature ventricular contractions (PVCs), ventricular and supra-ventricular tachycardia, atrial fibrillation. For monitoring and detecting such an irregular heart rhythm accurately, we are now developing a very cost-effective ECG monitor, which is implemented in 8-bit MCU with an efficient QRS detector using steep-slope algorithm and arrhythmia detection algorithm using a simple heart rate variability (HRV) parameter. This work shows the results of evaluating the real-time steep-slope algorithm using MIT-BIH Arrhythmia Database. The performance of this algorithm has 99.72% of sensitivity and 99.19% of positive predictivity. We then show the preliminary results of arrhythmia detection using various types of normal and abnormal ECGs from an ECG simulator. The result is, 18 of 20 ECG test signals were correctly detected.

Bedside patient monitoring by NFC

Abstract: The application of NFC for medical data collection is introduced in this research in order to reduce the process of vital signs measurement including heart rate, blood pressure and body temperature. By deploying a smartphone, all measured signals and records can be collected automatically via RFID chip from patient's beds. The proposed system is controlled from the smartphone through Android SDK by transmitting radio frequency to the target RFID chip and communicates with the microcontroller via RS232 serial port. All data are retrieved back and displayed in the smartphone. The microcontroller has EEPROM memory for recording all measured results continuously. This system improves the patient data collection process by reducing time, increasing the accuracy and easy to use. The results show that the percent of accuracy is about 90%. The application can show result correct and successful.

Analysis system for urine strip test using image processing technique

Abstract: Urine is waste that must be discharged from living bodies. In the medical field, urine is very useful for analyzing and curing diseases. The kidneys extract waste from the blood. Urinalysis is used to determine and inspect urinary function and system. This research developed an analytical System for a Urine Strip Test Using Image Processing to compare among human eyes and Urine automated with 100 samples. The result shows there is no differences with significant alpha below 0.05. In conclusion, any medical technician can use this system instead of a conventional method. It is cheaper and faster.

Time-frequency analysis of red-green visual flickers based on steady-state visual evoked potential recording

Abstract: The study of brain activity can be done by visual stimulus flickering at specific frequencies, Steady-State Visual Evoked Potential or as known as SSVEP. SSVEP is to stimulate the EEG signal to locate the desired goal of the experiment when a visual stimulus flickering with different constant frequencies and same duration. We aim to present a case study of electroencephalogram (EEG) signal by analyzing the frequency response of the red and green light. The stimulation is based on SSVEP by dividing the trial into two trials: single light and two lights. We considered three parameters that are light color, frequency and epoch interval. The optimal experimental results showed the classification accuracy rate of 74% and 75% for single and two color lights, respectively. The results can be considerably applied to the brain-computer interface (BCI) system.

Feature extraction of electrocardiogram signals using discrete sinc transform

Abstract: The most common cause of human death is cardiovascular disease (CVD) in today's world. Electrocardiogram (ECG) is widely used techniques for diagnosing CVD. Several methods have been proposed to classify arrhythmia of ECG. In this paper, a new method of extracting features from ECG signals is proposed. This method is built on completely modifying MFCCs (Mel-Frequency Cepstral Coefficients). No filterbank or hamming windows are used here for processing signals of ECG. The motivation is achieved by using a new transform for extracting features which is called discrete sinc transform (DSNT) that is used instead of DCT (discrete cosine transform). ECG signals are first filtered to remove noises and then normalized. Then energy of signals in the frequency domain is computed using Fast Fourier Transform (FFT). After that, discrete sinc transform is performed. Then, the 3rd derivative of these signals is calculated. Skewness, standard deviation, minimum, maximum and mean are calculated to extract 5 features of each ECG signal. Support vector machine (SVM) performs the classification of the extracted signals. Signals of ECG are collected from the MITBIH arrhythmia database. The obtained accuracy percent is 95.45% when classification is performed between normal and abnormal heart beats of ECG signals when signals of ECG are extracted using DSNT. The accuracy percent is 90.91% of classification when signals of ECG are extracted using DCT. So, DSNT is preferred in feature extraction of ECG signals.

Sleepiness classification by thoracic respiration using support vector machine

Abstract: It is widely known that many traffic accidents occur every year not only in Japan but also throughout the world. Sleepiness or drowsiness, which is the cause of dozing at the wheel, happens regardless of the physical condition of the driver at the time such as after having had meals or at midnight. This indicates that it is too difficult to expect the driver to avoid sleepiness or drowsiness by themselves. Therefore, various systems have been proposed to prevent traffic accidents caused by dozing at the wheel. In this study, we examined the relationship between subjective sleepiness during driving, which was evaluated by the Japanese version of the Karolinska sleepiness scale (KSS-J) and physiological parameters extracted from thoracic respiration signals. Then we tried to classify the existence of heavy, light, and no sleepiness using a support vector machine on those parameters. In this study, we determined a KSS-J score of 8 or 9, 6 to 8, and from 1 to 5 as signifying heavy, light, and no sleepiness states. The support vector machine was trained using three-quarters of the data for each subject and the remaining data was used as the testing data. This approach enabled us to obtain an accuracy of 89.4%. Therefore, it was suggested that thoracic respiration parameters were relevant to sleepiness.

Automated Quality Control for Proton Magnetic Resonance Spectroscopy Data Using Convex Non-negative Matrix Factorization

Abstract: Proton Magnetic Resonance Spectroscopy (1H MRS) has proven its diagnostic potential in a variety of conditions. However, MRS is not yet widely used in clinical routine because of the lack of experts on its diagnostic interpretation. Although data-based decision support systems exist to aid diagnosis, they often take for granted that the data is of good quality, which is not always the case in a real application context. Systems based on models built with bad quality data are likely to underperform in their decision support tasks. In this study, we propose a system to filter out such bad quality data. It is based on convex Non-Negative Matrix Factorization models, used as a dimensionality reduction procedure, and on the use of several classifiers to discriminate between good and bad quality data.

Design and construction of the hypoglycemia monito wireless system for diabetic

Abstract: Hypoglycemia event most found during intensification glycemia control, continuously monitoring blood glucose may decrease risk of hypoglycemia but invasive procedure and high cost. The aim of this study is design the novel system to monitor hypoglycemia system in diabetes patient to prevent severe hypoglycemia event and decrease frequency to check the blood glucose level of use Pathophysiology hypoglycemia event. Composition of the device was defined in 3 sensors of pulse rate variation, humidity variation and temperature variation via skin at the wrist By sending continuous data by WiFi module connected to WI-Fi-hotspot then all data were recorded on the website and analysis risk level and graph on a personal computer or application in smart phone real time display. Result of this process was measured hypoglycemic risk from 1 to 3 levels depend on temperature, humidity and pulse rate by compared with blood sugar from serum. This system is an additional choice to prevent severe hypoglycemia with non-invasive device. A Physician can access to data from tale-medicine device.

Development of an information system for medical equipment management in hospitals

Abstract: This research aimed to develop a WepMEt (Web application for medical equipment management in hospitals) employing the concepts of internet of thing, decision support system, and industrial engineering technique. The system was web-based and supported administrative tasks from both in-house and outsourcing contract service. The program was initially developed by rapid prototype-base method. Then working procedure was synthesized by specialists from Biomedical Instrumentation for Research and Development Center of Mahidol University, Medical Equipment Units of Rajavithi Hospital and Queen Sirikit National Institute of Child Health. The program incorporated 6 modules; 1) medical equipment registration 2) spare part registration 3) repairing and maintenance 4) preventive maintenance and calibration 5) medical equipment stock and 6) summary and report. The program had been tested for 3 months in a 500-bed hospital and an 800-bed hospital. The user satisfaction was evaluated. The evaluation result showed that the first 3 most satisfied indicators were; the operation meets the expectations. (4.11), meet the quality assessment system and overall performance (4.09), and the equipment information meets the user's requirement. (4.06).

Converting air quality monitoring low cost sensor data to digital value via mobile interface

Abstract: Particle pollution problem has become widely catastrophic in many countries around the world. During outdoor activity, many people have to face the effects of particle pollution in the atmosphere. This is the main cause for respiratory health defects, such as heart, vascular, stroke and lung disease. In order to protect and prevent people from the particle matter area, this research shows the step of implementing the mobile application sensor, while checking sensitivity when it is connected. Using this existing low cost SDS011 sensor with CE, RoHS and FCC standard and testing mobility detects PM 10 and PM 2.5 value. Moreover, this research plots linear regression graphs to present the trend of the PM that is parallel with the x-axis. This illustration shows the promising performance of the sensor application after only a short period of time collecting data. Not only is this device more convenient than the bigger model due to its small size, but its ability to connect to an app on smart phones makes it ubiquitous. Moreover, we look forward to our future work where multiple devices can be placed next each other in the same areas. This will lead us to make an accurate quarantine area of the air pollution problem. Furthermore, this data can be used in forecasting combined with respiratory, cardiovascular or related diseases.

Study of cybercrime and security in medical devices

Abstract: This paper presents a cybercrime in medical devices such as all of devices feed in RJ-45. The reason of the physician can monitoring the patient's data from medical devices in hospital and how to protect them using technique of anti-crime, we study behavior of hacker to fine history of his/her crime and understanding cybercrime method. Finally we use as forensic evidence by analyzing signal data from a database of DSI database crime comparison with cybercrime investigation. The result was that there was no security in medical devices procedure bad effects.

Application of ultrasound in determination gestation status in captive female freshwater stingray (potamotrygon sp.)

Abstract: Freshwater stingray is one of the popular ornamental fish trade in Thailand. Early pregnancy diagnosis could improve reproductive performance and management in captive breeding. Therefore, the main objective of this study was to apply and develop techniques for determination of gestation status in freshwater stingray by using ultrasonography. The stingrays were placed in a fitted tank with water in ventral recumbency position. The imaging was done in B-mode. The suitable frequency was 6.5 MHz, covered all size of pregnant female stingrays. However, the frequency may be converted to 5 MHz or 8 MHz to identify implanted embryo or trophonemata. The scanning showed 3 stages of gestation; first stage, mid stage and final stage with different characteristic. The present study suggested that this application, a noninvasive technique, can be continuously used for determination the gestation status from 4–12 weeks of gestational age in captive female stingray.

The development of biosignal processing system (BPS-SWU V1.0) for learning and research in biomedical engineering

Abstract: This paper presents a preliminary study on the development and the experiment of biosignal processing system (BPS-SWU v1.0) for learning and research application for undergraduate students in biomedical signal processing laboratory (BME LAB). The first version of our system, the frequency range of lung sound in the normal subjects was investigated by using spectral analysis technique. The experimental results indicate the characteristics of lung sound frequency in the form of spectrogram. The results show a frequency range of the lung sound in normal subjects from 34.65–119.40 Hz with certain frequency characteristics. The results are very promising.

HB-UWB transmission wavefrom model for wireless body area network

Abstract: This paper are presents characteristics of human body-ultra wide band (HB-UWB) transmission model for WBAN. The human body channel is more complex and distinctive than other application. This complex and irregular shapes that cannot be analyzed as a regular Therefore, to understand human body channel is so important. The channels model is classified with body mass index (BMI). The root raise cosine passband waveform is used as UWB transmitted waveform. The body channel transfer functions on human body are measured follow IEEE802.15.6 by using vector network analyzer (VNA) at frequency 3 GHz to 5 GHz. Distortion is important parameter for UWB waveform transmission. It can be fine from path loss model, which show compared of average power loss and peak power loss. The results can be applied to use evaluation channel model on human body.

Investigation on feasibility of using surface plasmons resonance (SPR) sensor for ultrasonic detection: A novel optical detection of ultrasonic waves

Abstract: Scientists and engineers have dreamed of a high resolution ultrasonic microscopic imaging, where the resolution of the ultrasound is required to be as high as optical resolution. In order to achieve this, of course, a very high frequency ultrasonic source in GHz regime is required as well as a highly sensitive ultrasonic camera in the same operating frequency range. In this talk, we will show some experimental results and discuss a feasibility and key issues of employing a standard Kretschmann based surface plasmons resonance (SPR) sensor system to perform high frequency ultrasonic imaging. At the end of the talk, we will discuss some ways to get around the issues.

The preliminary investigation of ear recognition using hybrid technique

Abstract: Ear recognition is one of the new patterns of biometrics. Ear structure is believed to contain specific and unique anatomical markers, which can be used both to distinguish it from others. In this paper, we derive preliminary of Ear identification based on the geometric features on 3D ear surface for 2D ear image. Principal components analysis (PCA) is used in this study for feature extraction to identify a person. The result is shown 92% recognition rate of 50 volunteers.

Performance characterization of aspheric polymer lens formed by gravity and surface tension: A high magnification portable microscope for smartphone and tablet

Abstract: An optical microscope is the primary equipment enabling us to see objects down to microscale level by human's naked eyes. In this talk, we discuss an approach to form an aspheric shape objective lens made of transparent adhesive polymer through gravity and surface tension. We tested imaging performance of such lenses and found that the total magnification exceeded 100x for instantly used liquid lenses, and was more than 450x for solid polymer lens. The shape of the lenses formed by the method is parabolic shape, which provides very good imaging performance.