ICACT20240402 Question.1
Questioner: cysy@bicol-u.edu.ph 2024-01-31 오후 5:02:06 |
ICACT20240402 Answer.1
Answer by Auhor sikandarshigri77@gmail.com 2024-01-31 오후 5:02:06
Chrome  Click!! |
| This is directed to Texture Analysis:
How does the GLCM texture analysis contribute to understanding gastric cancer stages?
Were there specific texture patterns or characteristics identified for early and advanced stages?
Thank you |
Basically, GLCM extracts information about spatial patterns and relationships among pixel intensities. Texture features play a vital role in determining diagnostic and classification tasks. however, most people ignore it and only focus on deep semantic features. Specifically GLCM feature helps analyze the different patterns of cancerous and noncancerous tissues in pathology images based on the different textures of cells. This was evident from our experiment that when we included GLCM features in our experiment the performance of our model improved significantly. |
ICACT20240402 Question.2
Questioner: cysy@bicol-u.edu.ph 2024-01-31 오후 5:07:53 |
ICACT20240402 Answer.2
Answer by Auhor sikandarshigri77@gmail.com 2024-01-31 오후 5:07:53
Chrome Click!! |
| Can you elaborate more on your ROC curve? thank you |
The ROC curve is one of the criteria to examine the performance of the model. Based on our experiment our model demonstrated an AUC of 96.7. The closer the curve is towards 100, the higher will be the performance of the model. It helps to evaluate how well the model separates the positive and negative classes. |
ICACT20240402 Question.3
Questioner: cysy@bicol-u.edu.ph 2024-01-31 오후 5:14:23 |
ICACT20240402 Answer.3
Answer by Auhor sikandarshigri77@gmail.com 2024-01-31 오후 5:14:23
Chrome Click!! |
| Were there any limitations or challenges encountered during the study, and how were they addressed? |
The small amount of data is one of the limitations of this study. In our future research, we use some data augmentation techniques and generate synthetic data to address this challenge and analyze the results. |
ICACT20240402 Question.4
Questioner: cysy@bicol-u.edu.ph 2024-01-31 오후 5:10:43 |
ICACT20240402 Answer.4
Answer by Auhor sikandarshigri77@gmail.com 2024-01-31 오후 5:10:43
Chrome Click!! |
| What are the potential future directions or extensions of this research? |
In the future, we are aiming to increase the number of data and also include more statistical features to enhance the performance. |
ICACT20240402 Question.5
Questioner: cysy@bicol-u.edu.ph 2024-01-31 오후 5:17:00 |
ICACT20240402 Answer.5
Answer by Auhor sikandarshigri77@gmail.com 2024-01-31 오후 5:17:00
Chrome Click!! |
| Are there plans to validate the proposed model on a larger dataset or in a clinical setting? |
In the future, we collect more data and get higher performance than the current performance of the model and will go for the clinical trial. |
ICACT20240402 Question.6
Questioner: ldaustero@bicol-u.edu.ph 2024-02-03 오후 12:45:59 |
ICACT20240402 Answer.6
Answer by Auhor sikandarshigri77@gmail.com 2024-02-03 오후 12:45:59
Chrome Click!! |
| How do you envision the integration of the proposed diagnostic tool into clinical practice? What potential impact could it have on improving the accuracy and efficiency of gastric cancer diagnosis and treatment planning? |
Well, this is our ongoing project with different other objectives like gastric cancer lymph node metastasis prediction and PD/WMD prediction etc. Pertaining to gastric cancer stages diagnosis, we will collect enough amount of data so that we can validate our model on a large amount of data thus enhancing the generalizability of the model. It will also go through external validation and will also be examined by expert pathologists. We may also need FDA approval to be integrated it into clinical practice. |
ICACT20240402 Question.7
Questioner: jllovido@bicol-u.edu.ph 2024-02-05 오전 2:33:12 |
ICACT20240402 Answer.7
Answer by Auhor sikandarshigri77@gmail.com 2024-02-05 오전 2:33:12
Chrome Click!! |
| Could you please provide further details on how data pre-processing was done? Were there specific data pre-processing challenges encountered?
|
|
ICACT20240402 Question.8
Questioner: jllovido@bicol-u.edu.ph 2024-02-05 오후 2:19:37 |
ICACT20240402 Answer.8
Answer by Auhor sikandarshigri77@gmail.com 2024-02-05 오후 2:19:37
Chrome Click!! |
| High-quality datasets for use as training data is a vital aspect of any machine learning model development. How did you ensure the accuracy and reliability of the medical images used in developing the classification model considering that a publicly available dataset was used?
|
|
ICACT20240402 Question.9
Questioner: madbrogada@bicol-u.edu.ph 2024-02-06 오후 2:04:28 |
ICACT20240402 Answer.9
Answer by Auhor sikandarshigri77@gmail.com 2024-02-06 오후 2:04:28
Chrome Click!! |
| Do you think it’s possible to take into account other factors like history of the patient as an improvement of this classification tool? |
|
