Facial Paralysis Assistant: FaCiPa

Hello everyone, as I promised you before, I’m here to introduce FaCiPa. I will introduce you to the details of how to make an application from scratch by approaching the FaCiPa application step by step, which you have come across in many interviews, interviews, and practices. Excuse my excitement today, because I feel like every project I do is my child. So much so that this project is very valuable to me, as it also contains memories from my own life.

As a result of an illness I had about 3 years ago, I had to go through a difficult and exhausting process of treatment. I did not know the name of this disease as I have mentioned it many times before. I’ll leave a little information below to make you aware of this. Please let us be informed. Because believe me, you don’t know the possibilities that life will bring you.

Facial paralysis occurs when facial nerve fibers from the brain to the face are removed due to certain causes (wind, heat exchange, virus, etc.) is a condition of loss of movement in the facial mimic muscles as a result of injury. Movements such as raising eyebrows, closing eyes, showing teeth, laughing and inflating cheeks are disrupted.

Movement (motor) orders from our brain are transmitted to our facial muscles through the facial nerve, allowing us to make the movements we want. If there is a disease or problem in the brain or facial nerve, and these disorders block the incoming message, facial paralysis occurs, so facial movements partially or completely disappear. Through this project, it is aimed to minimize damage by allowing the user to diagnose early before the stages of the disease progress further.

Facial nerve branching
Stages and Classification of Facial Paralysis

According to statistics and literature:

  1. First Degree Injury: the facial nerve is only stunned, the patient returns to normal life within eight weeks.
  2. Second Degree Injury: the facial nerve is damaged, but the outer part still remains intact. It shows the first signs of recovery after four months.
  3. Third Degree Injury: the facial nerve is severely damaged. Recovery is very slow and sometimes surgical intervention may be required.

According to this information and results, my project aims to enable individuals at risk of developing facial paralysis to be able to diagnose it at a certain rate by installing the app from their mobile devices without the help of an expert. FaCiPa is a mobile application, supported by the necessary image processing techniques, and is a platform where the methods to be used in the diagnosis of facial paralysis (raising eyebrows, closing and opening eyes, whistling, smiling by showing teeth) are integrated. The name of the app consists of certain letters of Facial paralysis, the equivalent of the disease in English, which is considered a universal language so that it can appeal to everyone.

Example of a person who has had and doesn’t have facial paralysis

The face is one of the important beings in our lives. Face-to-face communication includes not only words, but also gestures, body language, facial expressions and eye contact. This type of communication has been the most effective human interaction compared to other types of communication. Individuals with facial paralysis have asymmetry in their face, difficulty eating, drinking, speaking, or pronouncing sounds. In addition, this discomfort also leads to physical function and psychosocial problems due to their physical appearance . Such diseases negatively affect the social lives of individuals. As a result, individuals who experience facial paralysis can face psychological, physical and social disabilities every day.

What Are We Investigating?

‘Do you know facial paralysis?‘, which resulted in the participation of 250 people in the literature survey for the project we carried out as a result of our survey, it was observed that 63.2% of individuals did not have the ability to diagnose without expert help.

  • How are the place of facial paralysis in our society and the attitude of individuals on this issue?
  • If facial paralysis could be diagnosed as early as possible, how would the success rate in his recovery and the change in the person’s level of psychosocial relaxation be observed?
  • What is the ability of individuals in our society to diagnose without expert help?
The Blue distribution represents YES, the Red Distribution represents NO

Image processing is a method developed to convert an image into a digital form and perform some operations, to obtain a specific image or extract some useful information from it. The input of this method is an image, such as a video cross-section or a photo. The output corresponds to the desired or required part of the image. Usually, the image processing system treats images as two-dimensional signals when applying predetermined signal processing methods.

What Is The Purpose Of Image Processing Technologies ?

The purpose of image processing is divided into 5 groups.

  1. Visualization: Observing hard-to-see objects
  2. Image sharpening and restoration: Improve noisy images
  3. Image retrieval: Engaging and high-resolution image search
  4. Pattern Recognition: Identifying various objects in an image.
  5. Image recognition: Distinguish objects in an image.
The Place Of Server Communication In Technology

An API (Application Programming Interface) is an interface created so that functions belonging to one application can also be used in another application Dec. An API is a software agent that allows two applications to talk to each other. In other words, it is the messenger who requests your request to your provider and then sends you a response. The API defines application-independent functions that allow applications and definitions to change without compromising each other.

Since it is not possible to develop the Python language in the Android Studio environment, service will be written to provide communication between OpenCV, the image processing library, and the mobile base, and communication will be established through the server. The framework to be used is Angular-based Ionic Cordova, which is a hybrid application roof, in order to develop mobile applications. Ionic, Angular.it is built on the JS library, and the structures used in code development are also Angular.it is owned by JS. Open source, Ionic is home to thousands of mobile apps with thousands of followers and supporters. The Ionic Framework is built on Cordova. Cordova provides access to the hardware and system resources of the mobile device.

In the project, a mobile application was developed using image processing techniques. Because it is not possible to develop the Python language in the Android Studio environment, the image processing library OpenCV and the mobile base will provide communication through the server by writing a service.


Facial recognition is a biometric software application that can uniquely identify or validate a person by comparing and analyzing patterns based on a person’s facial lines. The facial recognition system is based on a high-resolution digital image. First, face Detection programs work to find and frame human faces in the image and ‘normalize’ the faces, which means that it puts the images in as full a front-facing format as possible, sizes and adjust the contrast with light. Then artificial intelligence software comes into play, creating a fingerprint-like face print (Faceprint).

Extracting facial images with OpenCV-DLIB Library

Face marks are used to tag and identify key face attributes in the image (source). Detecting facial marks is a subset of the shape prediction problem . An input image is handled and the shape determinant tries to determine the key points of interest along the shape.

Model of face taken with 68 coordinates

The operating logic of the Facial Landmarks Detection method in the Dlib library is used; a training set consisting of images of facial marks is used. These images are added to the new image by specifying the specific (x, y) coordinates of the regions surrounding each face structure. As a result, Facial Landmarks are a detection method that can be used to detect facial signs with high-quality estimates.

Extraction of facial images in a person with facial paralysis

As shown below, the human face is divided into a total of 68 coordinates, and a certain range of coordinates is applied for certain areas on the face.

In my next article, I will tell you about mobile application and server interaction. Happy today if I could enlighten you even a little. Hope to see you. Have a healthy day ✨


  1. https://www.medicalpark.com.tr/yuz-felci-nedir/hg-151
  2. https://en.wikipedia.org/wiki/Facial_nerve.
  3. ] Josef Finsterer (2008). Management of peripheral facial nerve palsy. Eur. Arch. Otorhinolaryngol, 265: 743-752.
  4. fizyoglik.com/yüz-felci-ve-egzersiz/
  5. Yüz Felci (Fasiyal Paralizi) başlıklı makalenin tüm hakları yazarı Prof.Dr.Selçuk ONART’a aittir.
  6. “Anatomi Atlası”, Michael Schuenke, Erik Schulte, Udo Schumacher, Palme Yayıncılık
  7. R. L. Delinger, J. M. VanSwearingen, J. F. Cohn, K. L. Schmidt, “Puckering and Blowing Facial Expressions in People With Facial Movement Disorders,” J. Phys Ther, vol. 88, pp. 909-915, August 2008.
  8. W.S.W Samsudin, K. Sundaraj, AI-Rehab Research Group, Universiti Malaysia Perlis (UniMAP)“Image Processing on Facial Paralysis for Facial Rehabilitation System: A review”, 23 – 25 Nov 2012.
  9. Mahdi Rezaei, Department of Computer Science, the University of Auckland, “Creating a Cascade of Haar-Like Classifiers: Step by Step”
  10. https://commons.wikimedia.org/wiki/File:Ionic_Logo.svg
  11. https://www.pyimagesearch.com/2017/04/03/facial-landmarks-dlib-opencv-python/
  12. http://www.microbiologynutsandbolts.co.uk/the-bug-blog/ringing-the-infection-bell-about-facial-nerve-palsy.

Leave a Reply