Mexican students have developed an early detection tool for lung cancer

Through image processing and pattern recognition techniques, Mexican students have developed a computer tool who performs a prediagnostic to identify patients with early-stage lung cancer, reported Sunday the National Polytechnic Institute (IPN).

The system Analyze images of computed tomography of the chest and clbadify them to prepare a pre-diagnosis, which can help medical specialists to develop their conclusive diagnoses.

For digital image badysis, young people use pattern recognition techniques (artificial neural networks), with which they clbadify images according to certain characteristics, characteristics and desired characteristics.

In Mexico, lung cancer is considered the second leading cause of death among men and the fourth leading to women.

The system was created by Ximena Cortes, Isaac Aguirre and Sergio Martínez Ávila, students from the IPN's School of Computer Science.

During its development, students worked with an image bank of 1,400 patients (spread over the Internet) via the site: "Collection of images from the Consortium of Lung Images Databases. (LIDC-IDRI) ", in the United States.

With the images, they have "formed" a series of algorithms able to recognize the differences between CT scans of healthy patients and those with pulmonary conditions.

Aguirre explained that they used some of the images to form the algorithm, another for the evaluation process to measure system performance and another fraction of the tomography was to perform tests with different images of the formation.

"Our system has achieved 95% accuracy and 95% completeness, which is the percentage of properly ranked pre-diagnostics."explained the student.

At the same time, Cortés pointed out that the interpretation of the image of a chest tomography "is not simple, because of the amount of information contained, which can make decision-making difficult and lead to misdiagnosis. "

While Martinez has stated that this computational tool does not claim in any way to substitute for clinical badysis and traditional methods, but prevent healthy patients from undergoing biopsies or unnecessary invasive treatmentss.

"The program is able to identify images that have the slightest indication of cancer, which allows the specialist to order a biopsy only for patients selected by the system," he said.

According to the National Cancer Institute of the United States, 70% of cancer diagnoses in early clinical conditions have a better response to treatment.