This readme file was generated on [2024-02-28] by [Weston Ross]

-------------------
GENERAL INFORMATION
-------------------

Title of Dataset:
Data from: Machine Learning Approaches in Non-Contact Autofluorescence Spectrum Classification;

Author Contact Information (Name, Institution, Email, ORCID)

	Principal Investigator: Weston Ross
	Institution: Duke University
	Email: weston.ross@duke.edu
	ORCID: 0009-0002-6214-2189

	Associate or Co-investigator:
	Institution:
	Email:
	ORCID: 

	Alternate Contact(s): Ashutosh Raman

Date of data collection: 20210817 - 20210830

Funding and grant numbers (if applicable): NIH EB030982

--------------------
DATA & FILE OVERVIEW
--------------------

File list (filenames, directory structure (for zipped files) and brief description of all data files):

This dataset contains spectral data for healthy (muscle) and tumorous (sarcoma) tissue when excited by a 405nm laser using a non-contact acquisition system.  The folder 'Spectral Data' contains two datasets - 'Sample Set 1' and 'Sample Set 2' - collected on 20210817 and 20210830, respectively.  Within each set are subfolders labelled for the type of tissue indexed by the mouse number for that dataset. E.g. Within 'Sample Set 2', the subfolder Healthy_Mouse_2 indicates the data in that folder are taken on healthy tissue (muscle) on the 2nd mouse of the dataset.  Within the subfolders are the CSV files that contain the spectral data.  See each specific README file within the dataset folders for an explanation of how these spectral data are organized and how the filenames indicate information about the data.

--------------------------
METHODOLOGICAL INFORMATION
--------------------------

Description of methods used for collection/generation of data: 

Spectral data were collected from recently sacrificed mice bearing sarcoma tumors. The device uses a standard epifluorescence design with a 180mW 405 nm laser to excite the tissue, and the emission is recorded using a CCD spectrometer.  This device has been previously described in [Tucker The laser spot is focused to a 1/e² spot size of .75mm at a working distance of 17mm. Multiple samples were taken from each mouse at locations on the tissue such that the adjacent acquisition points were not overlapping.

The mouse genotype used was LSL-KrasG12D/+; p53Flox/Flox and has been previously described [2]. Mice were obtained from Jackson Laboratory (Bar Harbor, ME). Soft tissue sarcomas were generated in the proximal portion of the medial or lateral gastrocnemius muscle as previously described [2].  Mice were manually examined for evidence of sarcoma growth or for likelihood of imminent morbidity. Once sarcoma tumors reached a verified size of 500 mm³, mice were removed from observation and prepared for sacrifice and tissue excision.

[1] https://ieeexplore.ieee.org/abstract/document/9492317
[2] https://www.nature.com/articles/nm1602


--------------------------
DATA-SPECIFIC INFORMATION <Create sections for EACH data file or set, as appropriate>
--------------------------

Sample Set 1:

File Naming Convention for Sample Set 1:
Name: [Mouse Number]-[Tissue Type][Sample Num.]-[Diode Current]-[Integration Time].csv

e.g. ‘mouse4-tumor9-250-100.csv’ indicates this spectra was collected on Mouse 4, the location was within tumor, it was the 9th sample collected from Mouse 4, the current of the laser diode was 250mA, and the integration time was 100ms.

File Structure:
Column A - Wavelengths Captured (nanometers)
Column B - Intensity Values

Missing data treatments: N/A

__

Sample Set 2:

File Naming Convention for Sample Set 2:
Name: [data type]-[Mouse Number]-[X Position]-[Y Position]-[Diode Current]-1.csv

e.g. ‘datatumor-mouse2-2.000-5.000-250-1.csv’ indicates this spectra was collected within tumor, on Mouse number 2 (of this specific sample set), the scan occurred at x = 2.0mm and y=5.0mm w.r.t. the initial scan location, the current of the laser diode was 250mA.  Integration time for all scans was 100ms and is not indicated in their filenames.

File Structure:
Column A - Wavelengths Captured (nanometers)
Column B - Intensity Values

   
Missing data treatments: N/A

-------------------------
USE and ACCESS INFORMATION <to be completed by RDR staff based upon information entered in metadata deposit form>
--------------------------

Data License: http://creativecommons.org/publicdomain/zero/1.0/

To cite the data: Raman, A., Zachem, T., Plumlee, S., Park, C., Eward, W., Codd, P., & Ross, W. (2024).Data from: Machine Learning Approaches in Non-Contact Autofluorescence Spectrum Classification. Duke Research Data Repository. https://doi.org/10.7924/r4vt1vh11