# Latvian herbal medicines under the infrared lens: An FTIR-ATR dataset

## TL;DR

This paper presents a dataset of infrared spectral fingerprints for various Latvian medicinal plants, enabling identification and quality control.

## Contribution

The novel contribution is a comprehensive FTIR-ATR dataset for Latvian medicinal plants, facilitating chemical analysis and species classification.

## Key findings

- The dataset includes 329 spectra from 19 medicinal plant species, suitable for chemical characterization and quality control.
- Minimal sample preparation was used, involving only grinding the plants into fine powder.
- The dataset supports multivariate analysis and chemometrics for spectroscopic data handling.

## Abstract

As more individuals use plant-derived products as prophylactic and therapeutic remedies, the herbal medicine market continues to grow rapidly. This trend shows the need for more advanced analytical methods, such as spectral “fingerprinting” – creating spectrum with a distinctive pattern, which is further used as a unique identifier for a plant.

The dataset was generated using Fourier-transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR) to produce distinctive spectral fingerprints of Latvian medicinal plants: red and white clover, purple coneflower, cowslip, common jasmine, sandy everlasting, horse chestnut, pot marigold, chamomile, coltsfoot, common daisy, elderberry, cornflower, Jerusalem artichoke, fireweed, common and Midland hawthorn, small-leaved and large-leaved lime, yarrow, and meadowsweet.

The samples were obtained from ten Latvian herbal medicine producers, along with one additional sample collected independently by Riga Stradiņš University researchers. The analytical method requires minimal sample preparation, involving only the grinding of medicinal plants into fine powder.

The research yielded 329 spectra, each corresponding to a distinct species of medicinal plant, which are sorted in 19 CSV files. These files contain spectral data suitable for various applications, including chemical characterization, species classification, and quality control. In addition, the dataset can be used for multivariate data analysis or chemometrics and will allow for signal processing and spectroscopic data handling.