• Currently 3.72/5
  • 1
  • 1
  • 2
  • 2
  • 3
  • 3
  • 4
  • 4
  • 5
  • 5

Rating: 3.7/5 (18 votes cast)

Authors Institution
Dana Alotaibi
Fatimah Alsafwani
Meernah Alabdullah
Mona Alhawaj
Noor Alhawaj
Supervised by: Dr. Mahbubnnabi Tamal
Co-Supervised by: Eng.Maram Alqarni
Imam Abdulrahman Bin Faisal University
Theme
Biomedical Engineering
Smart phone: a cost-effective point-of-care (POC) medical device for non-invasive diagnosis of anemia.
Background
  • Anemia defined as the lack of hemoglobin in the body and according to World Health Organization (WHO) [1] in 2005 stood 1.62 billion people have anemia worldwide.
  • Patients usually need to visit health care clinics regularly to provide blood through the puncture of the skin which is invasive and painful.
  • A point-of-care (POC) device using smartphone camera sensor are considered acceptable and accurate tools for measuring hemoglobin levels to detect anemia [2,3], where the patient can monitor blood condition without going to the hospital.
  • One point-of-care application for measuring hemoglobin level by smartphone cameras non-invasively is HemaApp [4]. However, one limitation is that different phone models use different camera specifications, which requires calibration to get accurate results.
Summary of Work

The  aim of this project is to design a new smartphone-based system (a smartphone app.) which will allow to detect anemia noninvasively. This project is devided into four main parts:

1- Design a calibration system to remove the variability in colorimetric measurements across different models of smart phones by reflecting RBG LEDs on RGB paper sequentially using MATLAB platform. A hardware system will be designed to minimize the ambient light effects along with other effects (e.g., light scattering, shading, and inappropriate focusing etc.) during the calibration procedure.

 2- Design  an interface circuit using Arduino Uno that will control the red, green and blue LED lights which will be directed toward finger tips.

 3- Capture videos of the reflected lights from the finger tips by the CMOS (complementary metal oxide semiconductor) image sensors embedded in the smartphone camera [5].

 4- Develop an algorithm to correlate the colorimetric values of the reflected monochromatic lights with the blood hemoglobin level.  

A smartphone-based app will be developed to integrate these four parts together in one system shown in (figure.1).

 

 

Conclusion

The attempt to reach accurate calibration factors were not successfully acquired. However, there are limitations that will be addressed in the future. More images should be taken in order to find the mean and standard deviation for each RGB channel to calculate more accurate calibration factors. Advanced components will be used to find the most accurate calibration factors (e.g. IR filter, diffuser and color passport). Moreover, an unknown image will be taken to examine the calibration factors for each RGB channel. A phone application will be developed to measure hemoglobin level after calibration.

Acknowledgement

we would like to thank everyone who had contributed to the successful completion of this project. we would like to express our gratitude to our research supervisor, Dr. Mahbubnnabi Tamal and Eng.Maram Alqarni for their invaluable advice, guidance and their enormous patience throughout the development of the research.

In addition, we would also like to express our gratitude to our loving parent and friends who had helped and given us encouragement

References
  1. B. d. Benoist, E. McLean, I. Egll, and M. Cogswell, "Worldwide prevalence of anaemia 1993-2005: WHO global database on anaemia," Worldwide prevalence of anaemia 1993-2005: WHO global database on anaemia., 2008.
  2. J. Punter-Villagrasa, J. Cid, C. Páez-Avilés, I. Rodríguez-Villarreal, E. Juanola-Feliu, J. Colomer-Farrarons, and P. L. Miribel-Català, "An Instantaneous Low-Cost Point-of-Care Anemia Detection Device," Sensors, vol. 15, pp. 4564-4577, 2015. 

  3.  S. Prasad and B. Roy, "Digital photography in medicine," Journal of postgraduate medicine, vol. 49, p. 332, 2003. 

  4. E. J. Wang, W. Li, D. Hawkins, T. Gernsheimer, C. Norby-Slycord, and S. N. Patel, "Hemaapp: Noninvasive blood screening of hemoglobin using smartphone cameras," in Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, 2016, pp. 593-604. 

  5. S. D. Kim, Y. Koo, and Y. Yun, "A Smartphone-Based Automatic Measurement Method for Colorimetric pH Detection Using a Color Adaptation Algorithm," Sensors, vol. 17, p. 1604, 2017. 
7.
Summary of Results
  • According to the selective light reflection, a specific color channel tend to have the pixel count at the highest level (255) when a light source (LED) and surface (paper) colors matches the channel color.
  • In contrast, a specific color channel tend to have the pixel count at the lowest level (0) when one or both of the LED and paper colors does not matches the color channel.

Comparing calibration factors of Galaxy S4 and iPhone8 using red LED reflection.

  • The calibration factor of RRR (reflecting red LED on red paper, then extract the image red channel) was calculated by subtracting the minimum value of the measured red channel (78) from the corresponded red ideal value (255). Resulting in a calibration factor of +177 (as indicated at the iPhone 8 red bar in the red LED chart above).

 

  • The calibration factor of RRG was calculated by subtracting the maximum value of the measured green channel (12) from the corresponded green ideal value (0). Resulting in a calibration factor of -12 (as indicated at the iPhone 8 green bar in the red LED chart above).
  • These values (calibration factors) will be added to reach the ideal pixel counts in the any taken images by the two phones (iPhone 8 and Samsung Galaxy s4).

Comparing calibration factors of Galaxy S4 and iPhone8 using green LED reflection.

Comparing calibration factors of Galaxy S4 and iPhone8 using blue LED reflection.

  • To integrate the parts shown in figure 1, a smartphone-based app will be developed which will allow to detect anemia noninvasively.

 

  • Designing an application requires outlines including customer needs, list of requirements, acceptance criteria and graphical user interface (GUI) to simulate the application. A customer needs shows the user's demands such as pages, icons, links and buttons. A stakeholder's needs are established in list of requirements to solve user's problem.
Designed & Managed by Innovative Technology®
Background
Summary of Work
Conclusion
Acknowledgement
References
Summary of Results
Send ePoster Link