I am doing my PhD research titled "Fourier Transform Infrared (FTIR) Spectroscopy to Analyse Human Blood", and we are using healthy and type II diabetes blood samples as testing samples to acquire FTIR spectroscopy data and compare their spectra.
Machine learning plays a central role in our data analysis process. We are currently testing various algorithms and would be very grateful if you could collaborate with our team by sharing your raw dataset. This data will be instrumental not only in generating synthetic datasets but also in training and validating machine learning models critical to our study.
Thus, if you could provide us with the CSV files for this research, we, of course, will include you/your team as co-authors in the publications outputted from this study. Could you help us?
I would be happy to assist in the generation of synthetic datasets to support your study. However, to ensure the synthetic data is realistic and scientifically valid, it is essential to have access to the exact parameters used in your real dataset. This will allow for accurate comparison and meaningful validation.
If you could kindly share an outline of your case, including key variables, measurement conditions, and any preprocessing steps, I will do my best to generate suitable synthetic data aligned with your study requirements.
Dear Prof. Khadersab.
Many thanks for your reply.
I need a dataset to test our model and compare our results with those of other authors.
If you share a dataset for type 2 diabetes with us, this will help us a lot.
Or if you have a normal blood sample dataset, it will be useful as well.
Best Regards
Sir,
What is your opinion on the dataset found at this link: https://www.kaggle.com/datasets/alexteboul/diabetes-health-indicators-dataset
Let me know your thoughts on it...
And if you select me as a coauthor for your current paper, I am ready to write the literature review section for your current paper.
Best Regards,
Thanks for your willingness to support us. What we need are datasets for type 2 diabetes, using FTIR spectroscopy. The CSV file must include data about the intensity for each wavelength (Absorbance or transmission). If you have a similar to this it will help us.
If I create a synthetic dataset, would that be acceptable to you?
A sample is attached, for your review
Please share your email address with me so I can send you the dataset.@
Thanks
Hello,
Thank you for reaching out and for sharing details about your research — it sounds like a valuable and impactful study.
At the moment, I’m not in a position to share any raw FTIR spectroscopy datasets of human blood serum (healthy and type 2 diabetic), as I currently don’t have access to such data myself or it is restricted due to privacy and ethical guidelines.
However, I recommend checking repositories like:
- MetaboLights
- Zenodo
- Figshare
- Kaggle (for synthetic or related spectral data)
These platforms sometimes host datasets related to biomedical FTIR studies.
If I come across a suitable dataset in the future or manage to generate one that’s ethically shareable, I’ll be happy to get in touch.
Wishing you all the best with your PhD and research publications.
Can we discuss this, in an online meeting in MS Teams?
For the sample dataset,send by me to your email..
Dear Mr. Yash...
Many thanks for your reply.
I will check the sources you provided.
Many thanks
- Badges
- Science topic