#chemometrics — Public Fediverse posts

If you want hear me ramble on pre-processing techniques for #spectroscopy feel free to register to the FREE webinar

"Tikhonov regularization for spectra smoothing and baseline subtraction"

I have two sessions scheduled on Sept 21st and Sept 22nd. The first session is generally best for the Americas. The second session's time is more convenient for EMEA and APAC, but you're free to join either!

All details are on the registration page 👇

https://events.humanitix.com/nirpy-webinar_3

#MachineLearning #Python #Chemometrics

📣 Wednesday 15th July, at 16:00 CET, I'll be hosting a webinar presented by Dr Mercedes Bertotto (Wittenborg University):

"Bridging Model Validation and Analytical Validation in Spectroscopy: The Role of the Accuracy Profile"

The idea behind the talk is to highlight that validation of #chemometrics models alone is often not sufficient. In addition to classical model validation approaches, analytical validation, traditionally used in analytical chemistry and laboratory testing, should also be applied in chemometrics and #spectroscopy workflows to ensure true method performance and reliability.

Registration is free at this link👇
https://events.humanitix.com/nirpy-webinar-2

#MachineLearning

I'm presenting a webinar on
"Class modelling and outlier detection in NIR spectroscopy"

We have two sessions scheduled
🔸 Session 1 (best for the Americas): Jun 16th, 06:00 - 07:00 (UTC+10)
🔸 Session 2 (best for EMEA - APAC): Jun 16th, 18:00 - 19:00 (UTC+10)

Registration is free, but spots are limited.

Details and abstract are on the registration page 👇
https://www.trybooking.com/DMSTO

#NIR #spectroscopy #NearInfrared #Chemometrics #Python #MachineLearning

📣 I'm planning to organize an online series of webinars/workshops, focused on #chemometrics and #MachineLearning for #spectroscopy (using #Python )

I'd like it to be beginner-friendly and informal.

🆓 Free registration

🏛️ Tutorial and research talks (including guest speakers and workshop sessions)

📅 Monthly or fortnightly schedule

💻 Fully online

If it sound like you may be interested, please read more and register your interest here
https://nirpyresearch.com/nirpy-webinars/

#Webinar

If your #spectroscopy dataset is small, deterministic data subdivision into training and test sets may be the way to go 🔬

The SPXY algorithm is an extension of the Kennard-Stone method that selects training samples to maximize coverage of both spectra (X) and response variable (Y) at the same time.

Here's a primer, with a #Python implementation for #NIR spectroscopy.

https://nirpyresearch.com/spxy-algorithm/

#Chemometrics #MachineLearning

It took a while, but I'm finally back to writing my blog 😎

The first installment for 2026 is an easy introduction to calculating information #entropy for optical spectra (or for any signal, really).

In my blog, I focus on #data analysis (#chemometrics, machine learning) applied to optical and near-infrared #spectroscopy Smoothing, or denoising, is one of the most common steps to work with spectroscopy data, and information entropy can be used as a criterion to guide the smoothing process.

Better still, the entropy of the derivative of a signal can help with that, because it accounts for the shape of the signal more naturally.

Read more at https://nirpyresearch.com/information-entropy-spectra/

#MachineLearning #NIR #Physics

An #introduction: I am a doctoral candidate at Rutgers with a focus on using #spectroscopy, especially #VibrationalSpectroscopy, #Chemometrics and data tools to understand chemical reaction systems.
At home, I am interested in #homelab, #birdphotography, and #3dprinting. I used to have other interests, but the PhD consumed them. Whenever I post, it'll probably be small things for funsies and work I do in #Python and LaTeX.

Development and validation of a new method by MIR-FTIR and chemometrics for the early diagnosis of leprosy and evaluation of the treatment effect.
Chemometrics and Intelligent Laboratory Systems
Volume 254, 15 November 2024, 105248
https://doi.org/10.1016/j.chemolab.2024.105248
#infrared #chemometrics #leprosy

📚 New post from me | Genetic Algorithm for Wavelength Selection Using NumPy

TL;DR A simplified implementation of a genetic algorithm (GA) for wavelength selection, using only @numpy and @sklearn

📝 Key Points:

🔸 The basics, including population, fitness function, crossovers, and mutations.

🔸 A step-by-step implementation of the GA for wavelength selection, with clear Python code examples.

🔸An example using NIR spectroscopy data to demonstrate the algorithm's application in selecting optimal wavelengths for predicting soil properties.
Improved regression results when comparing the performance before and after optimization.

This post may be valuable for spectroscopists, chemometricians, and data scientists looking to optimise feature selection in #spectroscopy datasets using evolutionary algorithms.

🌐 Full post available here
https://nirpyresearch.com/genetic-algorithm-wavelength-selection-numpy/

#chemometrics #python #numpy #MachineLearning

New publication: a novel quantitative method to validate accurate scale-up of chemical reactions in real time.

Discover the full methodology at https://oa.eu/E4xb6X

#datascience #chemometrics #cmc #pharma

☕ Here's a bit of technical content from me - today a deep dive on #baseline correction methods.

📈 Baseline correction is a preprocessing technique to remove background signal and isolate peaks in hashtag#spectroscopy data.

📝 In my recent post I discuss two methods:
1. Wavelet transform (WT) - Decomposes signal into components at different frequencies. Lowest frequency component represents baseline and can be removed.
2. Asymmetric least squares (ALS) - Fits a smooth baseline function, penalising positive deviations more than negative ones.

TL;DR: WT method is intuitive but can distort peaks. ALS produces better results.

🔎 Both methods are applied on a #Raman spectrum and an X-ray fluorescence (#XRF) spectrum. ALS gives a cleaner baseline correction and it's effective for removing broad, slowly varying background while preserving sharper spectral features.

#chemometrics #Python #MachineLearning #wavelets #regression

https://nirpyresearch.com/two-methods-baseline-correction-spectral-data/

I had not see this #chemistry journal before: https://www.sciencedirect.com/journal/artificial-intelligence-chemistry That's a lot of #cheminformatics #chemometrics journals we have now. #elsevier

JEOL NMR users: you can now do chemometrics within Delta using the ChemoSpec package #chemometrics #NMR #NMRchat
https://chemospec.org/posts/2023-08-23-CS-Delta/CS-Delta.html

📯 New blog post: "
Wavelength selection with a genetic algorithm"

If you're interested in #spectroscopy, especially NIR or Raman, or #chemometrics, a suitable selection of wavelength bands is often needed to produce a good statistical model.

Genetic algorithms are optimisation procedures loosely inspired by the mechanism of evolution by natural selection.

In this point, I'm working through an example of variable selection with a genetic algorithm for regression.

#MachineLearning #GeneticAlgorithms
#NearInfrared

https://nirpyresearch.com/wavelength-selection-genetic-algorithm/

Training a #chemometrics or #MachineLearning model involves dividing a dataset into at least two groups: a calibration subset and a validation subset. The calibration set is used to fit the model, while the validation set is used to check that the trained model works as intended on a set of unknown data.

The split into calibration and validation can be done randomly or according to some predefined criterion.

One of the first, non-random, criteria for the subdivisions is called the Kennard-Stone algorithm and it's explained in my blog using #spectroscopy data.

https://nirpyresearch.com/kennard-stone-algorithm/

Wavelet decomposition is very popular in image analysis and processing. Here's my attempt to share some code to perform spectral smoothing (or denoising, to be fancy) using the same principle.

Wavelet denoising of spectra • NIRPY Research
https://nirpyresearch.com/wavelet-denoising-spectra/

#ImageProcessing #chemometrics #spectroscopy #wavelets #DataProcessing

PLS-Discriminant analysis in
#Python

#MachineLearning #chemometrics

iCite: "ITERATIVE RE‐WEIGHTED COVARIATES SELECTION FOR ROBUST FEATURE SELECTION MODELLING IN THE PRESENCE OF OUTLIERS (IRCOVSEL)"
Journal of Chemometrics 2022
https://doi.org/10.1002/cem.3458.
#openaccess #chemometrics #featureselection

Not every data point is created equal. Outliers in #NearInfrared #spectroscopy can be detected with PLS decomposition using #scikit_learn in #Python.

https://nirpyresearch.com/outliers-detection-pls-regression-nir-spectroscopy-python/

#chemometrics

Webinar:
"The contribution of the method of visualization “t-SNE” to the analysis of data from vibrational spectroscopy".
François Stevens (CRA-W, Gembloux, Belgium)

29 Nov 14.00 UTC.
Zoom: https://ucd-ie.zoom.us/j/67096694355

#chemometrics #foodscience

Greetings All! I'm a retired academic natural products chemist with extensive background in spectroscopy, chemometrics and software development in R. #rstats #nmrchat #chemometrics

#Introduction to #Mastodon community on new instance!

I'm an #academic interested in #metabolomics #chemometrics and #diseases. My group focuses on decoding #smell of #diseases and #scent in general. We work with #clinicians across the globe to build predictive #models. We are based in #Manchester, #UK.

If you are interested in #scientific #research, #academia, #coffee or #cricket, come say hi! Also, if you have any #tips and #tricks to find cool stuff on here, please #educate me!

#Introduction to #Mastodon community!

I'm an #academic interested in #metabolomics #chemometrics and #diseases. My group focuses on decoding #smell of #diseases and #scent in general. We work with #clinicians across the globe to build predictive #models. We are based in #Manchester, #UK.

If you are interested in #scientific #research, #academia, #coffee or #cricket, come say hi! Also, if you have any #tips and #tricks to find cool stuff on here, please #educate me!

Oh hey… adding some hashtags to my timeline. My interests as a dull list: #chemistry #bioinformatics #ai #ml #Proteindesign #biologics #epidemiology #immunology #aviation #stem #proteinstructure #biomanufacturing #software #chemometrics #stemjobs #drugdiscovery #drugmanufacturing