This work is an important contribution to the field of machine-learning-enabled NIR spectroscopy, providing a systematic approach to selecting representative subsamples from existing data using quality measures, diagnostic tools, and visualization techniques. provide researchers with a method.
Researchers from Graz University of Technology and Christ University have presented a systematic method for selecting representative subsamples from existing studies using an extensive set of quality measures and visualization strategies. In their article, published in AAPS PharmSciTech, Amrit Paudel and Gobi Ramasamy, describe a systematic and structured procedure for selecting subsamples from historical data (1). They offer a wide range of detailed quality measurements, diagnostic tools and visualization techniques.
This study included different doses in milligrams, different shapes and sizes of dosage forms, slotted tablets, three different manufacturing scales (laboratory, pilot and production), and different coatings (coated vs. uncoated). ), we used an open-source tablet data set consisting of ), more. The model was developed at one scale, and researchers investigated how well the top model applied when tested on new data, such as pilot scale and production (full) scale.
Researchers have demonstrated the choice of appropriate hyperparameters and their impact on the performance of artificial neural network multi-layer perceptron (ANN-MLP) models. We discuss hyperparameter tuning approaches and performance choices using available references for the data under investigation. The model extension from laboratory scale to pilot scale was successfully demonstrated.
ANN-MLP is a type of artificial neural network widely used for supervised learning. It is a feedforward neural network with multiple layers of neurons, including an input layer, one or more hidden layers, and an output layer. Each neuron in the network receives input from the previous layer, performs mathematical operations on the input, and then passes the result to the next layer. ANN-MLP is used for various applications such as database exploration, calibration modeling, image recognition, speech recognition, and natural language processing.
Near-infrared (NIR) spectroscopy is ideal because it is non-destructive and non-invasive, requires little or no sample preparation, and can significantly reduce overall analysis time. It becomes a real-time analysis tool. This technology is primarily used in the pharmaceutical, agriculture, food and dairy, cosmetics, pulp and paper, and precision medicine industries.
Derivatization, normalization, scatter correction, and advanced approaches are some of the data preprocessing techniques used to hide physical information and retrieve chemically relevant information from NIR data. Modeling is used after the physical/chemical information has been segmented. Principal Component Regression (PCR) and Partial Least Squares Regression (PLS) are used in multivariate linear models.
This work is an important contribution to the field of machine-learning-enabled NIR spectroscopy, providing a systematic approach to selecting representative subsamples from existing data using quality measures, diagnostic tools, and visualization techniques. provide researchers with a method. This work provides a framework for choosing appropriate hyperparameters and demonstrates extension of the model from laboratory scale to pilot scale.
reference
(1) Ali, H.; Muthudoss, P.; Ramalingam, M. Kanakaraji, L.; Podel, A.; Ramasamy, G. Machine learning enabled NIR spectroscopy. Part 2: Workflow for selecting a sample subset from publicly accessible data. AAPS PharmSciTech. 2023, twenty four, 34. https://link.springer.com/article/10.1208/s12249-022-02493-5
