Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Early detect larynx cancers using sound technology

No description
by

Ammar Alsamei

on 11 March 2015

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Early detect larynx cancers using sound technology

Early Detection of Larynx Cancer Using Sound Technology
Contribution
Contributing to start a new scientific field by taking advantages of engineering technologies to automate medical applications.
Proving the existence of sound differences between various larynx diseases that may affect sound , such as cancer.
Contributing to design of an entirely new system to simplify the cancer diagnosis .

Introduction
Goal
:
Building a strong and new software application to early detect and diagnosis Larynx Cancer through sound features.

Benefited
:
Lots of normal and larynx infected people.
Smokers(especially dark skin smokers) between the ages of 50 and 70.
Weak possibilities health centers and hospitals.
Literature Review – Topics
The larynx(or voice box ):
located in the throat .
Voice (or vocalization):
Produced by humans and other vertebrations using the lungs and the vocal folds in the larynx .
Generated by airflow from the lungs as the vocal folds vibrate .

Methodology
1- Collecting voice samples:
The process of collecting patients voice samples , detailed ,information and medical examination
.
Conditions:
Quiet room.
Special microphone.
4-6 cm space between mic and patients mouth.
Vowel recorded /a/ or /ann/ /a/.
Repeating the sound for 5 times
Supervisor
DR.AbdulQuddoos Saad Alsamei
Done by
Ammar Alsamei
Amr Mohammed
Buthina Ahmed
Hawa Shehab
Nawal Abdualghani
Saddam Taher
Problem Statement & Motivation
Development of new and advanced medical methods.
The ability to cure larynx cancers if detected in early stages.
Financial ,psychological and time difficulties .
Effect of larynx cancer on sound features.
Aims to develop a new technology used to early detect larynx cancers through sound.

Objectives
This project aims to early detect the larynx cancers in early stages by exploitation of sound technology ,and it is going to be achieved through:
Surveying the literature for understanding previous studies .
Finding , gathering and recording normal and abnormal sounds of patients .
Differentiating clearly between normal and abnormal sounds .
Exactly differentiating if abnormal sounds is just an injury in one of larynx disease, or it is a cancer .
Building an accurate and easy to use application software that perform this job .

Literature Review - Topics
Speech:
Is an immensely information-rich signal exploiting:
Frequency, amplitude, time modulated carriers .
Convey information .
Speech processing :
is the study of speech signal and the processing methods of these signals .

Literature Review - Topics
Disorders of the voice involve problems with:
Pitch is the highness or lowness of a sound based on the frequency of the sound waves.
Loudness is the perceived volume (or amplitude) of the sound.
Quality refers to the character or distinctive attributes of a sound.

Literature Review – Techniques
Algorithms and techniques used in this study:
Mel Frequency Cepstral Coefficient (MFCC).
Dynamic Time Warping algorithm (DTW).
Chebyshev filters.
Support Vector Machine(SVM).

Literature Review –Tools
Common tools used in this study:
Java language platform.
MATLAB platform.

Methodology
2. Noise Filtering and Feature Extraction:
Noise Filtering
The process to eliminate noise in the signal.
Feature Extraction
The process where we extract the features that distinguish one disease from the other.
Possible tool:
Chebyshev type2 Filter.

Methodology
3. Preparing the Training data set:
Classifying into groups based on the diseases.
Storing the training data set in the database.

Methodology
4. Signal Comparison:

Comparing the input signal with the training data set.

Possible algorithm:
Support Vector Machine(SVM).

Methodology
Results
Gathered and recorded normal and abnormal sounds.
Differentiated clearly between normal and abnormal sounds.
Differentiated if abnormal sounds is just an injury in one of larynx disease, or it is a cancer.
Built an accurate and easy to use application software that perform this job.

References
http://www.nidcd.nih.gov/.
http://en.wikipedia.org/.
Boyanov B, Chollet G: Pathological voice analysis using cepstra, bispectra and group delay functions. In Proceedings of International Conferenceon SpokenLanguage Processing, Alberta, Canada,vol. 2,1039-1042,12-16October, 1992.
K. Aiwarya Lakshmi: Analysis Of Larynx Disease using Dynamic warping Algorith. Is registered in International Journal of Advances in Computer Science and Technology, Volume 3, No.8, August 2014.
http://www.ifp.uiuc.edu/~minhdo/teaching/speaker_recognition.
http://www.idosi.org/mejsr/mejsr20(12)14/139.pdf
Full transcript