Using other feature extraction methods

Oct 13, 2025 by leolavanderia.com.br 22 views

Identification of Lung Cancer in Medical Image Chest X-Ray (CXR) using the Probabilistic Neural Network Method

Introduction

Lung cancer is one of the leading causes of death worldwide, with a high mortality rate due to late detection. The disease requires rapid and directed treatment to be effective. Usually, the diagnosis of lung cancer is done through medical image examination using a chest x-ray (CXR). However, the results of the examination can only be read and interpreted by medical experts, such as doctors. In conditions with many cases, the diagnosis process by a doctor requires a small amount of time and still allows interpretation errors.

The Importance of Early Detection

Early detection of lung cancer is crucial to improve patient prognosis. The current digital era has allowed us to use image analysis methods to identify various diseases, including lung cancer. One approach that is widely studied is the processing of digital images with the application of artificial neural networks (artificial neural networks) to help the disease identification process.

The Role of Artificial Neural Networks in Disease Identification

Artificial neural networks (ANNs) have been widely used in various fields, including image analysis. ANNs are inspired by the structure and function of the human brain and can learn from data to make predictions or decisions. In the context of disease identification, ANNs can be used to analyze medical images and identify patterns that may indicate the presence of a disease.

The Probabilistic Neural Network (PNN) Method

This study offers a Probabilistic Neural Network (PNN) method to identify normal images and images that show cancer from lung images. The identification process in this study consists of several important steps:

Pre-processing

The pre-processing step is done to prepare the image to be ready to be analyzed. This includes removal of noise and increasing image contrast. The goal of pre-processing is to enhance the quality of the image and remove any irrelevant information that may affect the analysis.

Segmentation

At this stage, the Canny Edge Detection method is used to detect the edges of the image, so that the suspected area contains cancer can be clearly separated from other parts. Segmentation is an important step in image analysis, as it allows us to focus on specific regions of interest and ignore irrelevant information.

Extraction of Features

Features of the image that have been segmented are extracted using the Gray Level Co-Octurrence Matrix (GLCM). This method allows us to get information about image texture that can provide an indication of the presence of cancer. The GLCM is a statistical method that analyzes the texture of an image by calculating the probability of different gray levels occurring together.

Classification

The extracted features are then used to classify the image as either normal or cancerous. The PNN method is effective in image classification into a normal category or cancer.

Results and Discussion

The results of this study indicate that the proposed method is able to identify the presence of lung cancer with accuracy reaching 93.33%. This number shows that the use of technology in medical image analysis can be a promising alternative to accelerate and increase the accuracy of cancer detection, which in turn can improve patient prognosis.

Conclusion

The application of probabilistic neural networks in processing CXR images offers new hopes in early detection of lung cancer. This can reduce dependence on doctor's assessment and increase accessibility and diagnosis speed, especially in areas with limited medical resources. Along with advances in technology and further research, it is hoped that this method can be widely implemented to save more lives through faster detection and treatment.

Future Research Directions

Further research is needed to improve the accuracy and robustness of the PNN method. This can be achieved by:

Collecting more data: Collecting more data on lung cancer images can help to improve the accuracy of the PNN method.
Improving the pre-processing step: Improving the pre-processing step can help to enhance the quality of the image and remove any irrelevant information that may affect the analysis.
Using other feature extraction methods: Using other feature extraction methods, such as texture analysis or wavelet analysis, can help to provide more information about the image and improve the accuracy of the PNN method.

Limitations of the Study

The study has several limitations, including:

Small sample size: The study used a small sample size of lung cancer images, which may not be representative of the larger population.
Limited data collection: The study only collected data on lung cancer images and did not collect data on other types of images.
Limited feature extraction methods: The study only used the GLCM method for feature extraction and did not use other methods, such as texture analysis or wavelet analysis.

Conclusion

In conclusion, the PNN method is a promising approach for identifying lung cancer in medical image CXR. The method has been shown to be effective in classifying images as either normal or cancerous, with an accuracy of 93.33%. Further research is needed to improve the accuracy and robustness of the PNN method, but the results of this study suggest that the method has the potential to be widely implemented in clinical settings.
Frequently Asked Questions (FAQs) about the Probabilistic Neural Network (PNN) Method for Lung Cancer Detection

Q: What is the Probabilistic Neural Network (PNN) method?

A: The PNN method is a type of artificial neural network that uses a probabilistic approach to classify images as either normal or cancerous. It is a machine learning algorithm that can be trained on a dataset of images to learn the patterns and features that are associated with lung cancer.

Q: How does the PNN method work?

A: The PNN method works by first pre-processing the image to remove noise and enhance the quality. Then, it uses the Canny Edge Detection method to detect the edges of the image and segment the image into different regions. The Gray Level Co-Octurrence Matrix (GLCM) method is used to extract features from the segmented image, and these features are then used to classify the image as either normal or cancerous.

Q: What are the advantages of the PNN method?

A: The PNN method has several advantages, including:

High accuracy: The PNN method has been shown to be highly accurate in classifying images as either normal or cancerous.
Robustness: The PNN method is robust to noise and other types of image degradation.
Flexibility: The PNN method can be used to classify images from different modalities, such as chest x-rays and computed tomography (CT) scans.

Q: What are the limitations of the PNN method?

A: The PNN method has several limitations, including:

Small sample size: The PNN method requires a large dataset of images to train the model, and the sample size may be limited in some cases.
Limited feature extraction methods: The PNN method uses the GLCM method for feature extraction, which may not be the most effective method for all types of images.
Computational complexity: The PNN method can be computationally intensive, especially for large datasets.

Q: Can the PNN method be used for other types of cancer detection?

A: Yes, the PNN method can be used for other types of cancer detection, such as breast cancer and colon cancer. However, the accuracy and robustness of the method may vary depending on the type of cancer and the modality of the image.

Q: How can the PNN method be implemented in clinical settings?

A: The PNN method can be implemented in clinical settings by:

Training the model: The model can be trained on a dataset of images from a clinical setting.
Testing the model: The model can be tested on a separate dataset of images to evaluate its performance.
Integrating the model: The model can be integrated into a clinical decision support system to provide real-time feedback to clinicians.

Q: What are the future directions for the PNN method?

A: The future directions for the PNN method include:

Improving the accuracy and robustness: The accuracy and robustness of the PNN method can be improved by using more advanced feature extraction methods and by incorporating more data from clinical settings.
Expanding to other types of cancer: The PNN method can be expanded to other types of cancer, such as breast cancer and colon cancer.
Developing a clinical decision support system: A clinical decision support system can be developed to integrate the PNN method with other clinical decision-making tools.

Q: What are the potential applications of the PNN method?

A: The potential applications of the PNN method include:

Early detection of lung cancer: The PNN method can be used to detect lung cancer at an early stage, which can improve patient outcomes.
Personalized medicine: The PNN method can be used to develop personalized treatment plans for patients with lung cancer.
Clinical decision support: The PNN method can be used to provide real-time feedback to clinicians to support clinical decision-making.