ICACT20250002 Question.1
Questioner: imst.ltd@gmail.com 2025-02-01 오후 10:11:20 |
ICACT20250002 Answer.1
Answer by Auhor jianms@nfu.edu.tw 2025-02-01 오후 10:11:20
Chrome  Click!! |
| It is a 3D cell culture system.
With different depth, how do you check the clarity of focus and define the scale? |
It is a good question.
Since it is 3D cell culture, differnt depth of the focus would cause the clarity of the image be different.
To measure the 3D cell, we suggest to use the gel with beads as the scale for measurement.
In other words, the known beads in the microscope image could be used as the scale for measurement. |
ICACT20250002 Question.2
Questioner: neil990006@gmail.com 2025-02-01 오후 10:15:37 |
ICACT20250002 Answer.2
Answer by Auhor jianms@nfu.edu.tw 2025-02-01 오후 10:15:37
Chrome Click!! |
| what kind of the image recognition is used in this research? |
Currently, we use the supervised learning as the AI training for the image recognition.
The unsupervised learning for the 3D image recognition is under development. |
ICACT20250002 Question.3
Questioner: lcf1024@mail.ntou.edu.tw 2025-02-02 오후 1:05:30 |
ICACT20250002 Answer.3
Answer by Auhor jianms@nfu.edu.tw 2025-02-02 오후 1:05:30
Chrome Click!! |
| 1. The authors purpose a unique Dynamic Bio-Information Image Recognition System (DBIRS) built for the real-time monitoring of cancer cell apoptosis produced by available medical therapy. Utilizing advances in microscopy, AI-based image recognition, and parallel computing across distributed edge devices, the proposed system dramatically boosts the precision and efficiency of cellular investigation. The system's efficiency is proven through trials on K-562 cells, where it attained a recognition accuracy of 97.41%. Please demonstrate the reasons for the recognition accuracy of 97.41% in detail. |
Thanks for your question.
According to the same image from the microscope, we would recognize the image (cells) by researchers' manual operation and AI recognition.
Due to the cell image would be cut by the edge of photo or crowded together,
the counting result of manual operation will be a little differnt from the recognition result of AI.
Hence, recognition accuracy can be estimated by AI recognization result to manual operation. |
ICACT20250002 Question.4
Questioner: lcf1024@mail.ntou.edu.tw 2025-02-02 오후 1:09:26 |
ICACT20250002 Answer.4
Answer by Auhor jianms@nfu.edu.tw 2025-02-02 오후 1:09:26
Chrome Click!! |
| Please compare the contributions of the proposed DBIRS, in detail. A detailed discussion about prior technologies is suggested to add extend journal paper in detail. Please elaborate the technique features of the proposed method in detail. |
Thanks for your suggestion.
Considering the shape of size of cell K562, the reconization based on the supervised learning is still available.
However, when the drug or medicine added, the status of cell would change.
Transition period of cell change should be defined in advance.
It is difficult to define the threshold for cell status whether it is alive or killed.
We are still under working for the definition and training the AI module. |
