NIRS Vision Tutorial – Instrument Operation 8.105.
Metrohm AG CH-9100 Herisau Switzerland Phone +41 71 353 85 85 Fax +41 71 353 89 01 info@metrohm.com www.metrohm.com NIRS Vision Tutorial – Instrument Operation 8.105.
Technical Documentation Metrohm AG CH-9100 Herisau teachware@metrohm.com This documentation is protected by copyright. All rights reserved. Although all the information given in this documentation has been checked with great care, errors cannot be entirely excluded. Should you notice any mistakes please send us your comments using the address given above.
Table of Contents Table of Contents Table of Contents III 1 1 Introduction 1.1 Prerequisites for this Tutorial........................................................... 1 1.2 Symbols and conventions ................................................................ 1 1.3 Field and parameters ....................................................................... 1 2 Getting Started 3 2.1 Starting Vision ...................................................................
Table of Contents 6 Routine Analysis 28 1.8 Setting up an Operations Method .................................................. 28 1.9 Analyzing Stored Data.................................................................... 31 1.10 Recalling Stored Results ................................................................. 32 7 7.1 IV Appendix 33 21 CFR Compliant Mode.................................................................
1.1 Prerequisites for this Tutorial 1 Introduction NOTE This tutorial describes the use of the Vision Software, it does not include specialties of the Vison Pharma version. 1.1 Prerequisites for this Tutorial In order to be able to follow this tutorial, the following prerequisites must be met: The Vision Software must be installed on your computer. Information on how to install Vision can be found in the document “Vision installation” (8.108.8006EN) delivered with the software.
1 Introduction Do not fill in fields or change parameters not mentioned in the following instructions. Explanations for all parameters can be found in the Vision Reference manual (8.105.80010).
2.1 Starting Vision 2 Getting Started 2.1 Starting Vision Once the software is installed, a Vision Icon can be found on the desktop of your computer. 1. Double click on 2. In the Login dialog box, enter the following default login information: . User ID NIRS Password NIRS 3. Press [Enter] or click on [OK]. 2.2 Setting-up a Project In Vision, a Project must be set up, before any data (instrument diagnostics or spectral data) can be collected.
3. 2 Getting Started Back in the Configure Data Sources window, click on [OK] Connect to the instrument 1. In the Connect to Instrument dialog box Select Acquire New Data The Select Data Collection Method dialog box appears. Before the Data Collection Method can be defined, project options and system options must be established. 2. In the Select Data Collection Method dialog box Click on [Cancel]. 3. If asked “Turn Lamp Off?”, click on [No]. 2.
2.5 Instrument Setup Diagnostics model, module and amplifier type, number of scans) and selected performance test parameters (like pass/fail limits). Connect the instrument 1. Click on Acquire Connect Create Data Collection Method 1. In the Select Data Collection Method dialog box click on [New] 2. In the Edit Data Collection Method dialog box enter a name for the DCM. We recommend the following structure for DCM naming: “name of instrument” – “detector type”.
3 Acquiring Spectral Data 3 Acquiring Spectral Data At this point you should have performed the necessary instrument diagnostic tests and be in your Project “tutorial”. If you are not: Open Project „tutorial“ 1. Click on File Project Open 2. In the Open Project dialog box, select “tutorial” click on [Open] 3. Click on Acquire Connect 4. In the Select Data Collection Method dialog box, Select your method e.g.
3.1 Accessing Tutorial Data 4. If asked “The product ID Test Samples does not exist. Do you wish to create a new product?” click on [YES]. 5. In the Add New Product dialog box enter the following information: Product Description Click on [OK]. e.g. Enter today’s date. Scan other samples 1. Repeat the steps in the sections ”Scan and save samples“ and “Entering Constituent Values“ until your entire test samples have been scanned and saved. 2.
3 Acquiring Spectral Data Restore Tutorial Data project 1. Click on Mode Data Acquisition 2. Click on File Restore Restore Project 3. In the Select project zip file to restore dialog box, go to the folder C:\vision\Tutorial Data\TutProj Double click on open the file xdsstart.zip In the Preview project to be restored dialog box click on [OK] The Project xdsstart will be unzipped and restored automatically to your Vision software. 4.
1.1 Creating a Library 4 Qualitative Model Development A qualitative model consists of a library of spectra of known products. A spectrum of an unknown product can then be compared with the spectra in this library. If a matching spectrum is found in the library, the unknown product can be identified. The following steps are required for developing a qualitative model: 1. Creating a library 2. Sample Selection 3. Developing an Identification Method 4.
4 Qualitative Model Development Open Project “xdsstart” 1. Click on the + sign next to the green cuvette icon. Add Product to the Library 1. Drag the product “AcDiSol” from the products folder into the library folder. 2. In the Library Product dialog box confirm the product ID and click on [OK]. The product “AcDiSol” now appears under the library name next to a gray (empty) cuvette.
1.2 Sample Selection Vision now performs calculations according to the parameters defined in the previous steps. Figure 1 Result Sample Selection for qualitative Analysis 1 Math treated spectra 2 Frequency versus Probability 3 Sample location in 2D Principal Component Space 4 Explained Variance Color Code: Dark blue = Training Set; samples determined to be suitable for library development.
4 Qualitative Model Development Include other Products 1. Repeat the Sample Selection Process for all the products in the xdsstart project. 1.3 Developing an Identification method An Identification Method is required to differentiate products which differ in chemical and/or physical properties. Develop an Identification Method 1. Click on Mode Qualitative Analysis Identify Method Development 2. File Library Open, select the library xdslib 3.
1.3 Developing an Identification method 7. In the Library Method dialog box optionally enter the following information: Comments “Training and today’s date”. Click [OK]. Vision now computes the required information on all spectra for all products in the library. It automatically analyzes each spectrum in the training sets as if it were an unknown to determine if the method is capable of identifying it.
4 Qualitative Model Development Save Results 1. Click on Identify Save Results 2. If asked “Save identify method #1 as library global identify method?” click on [YES] 1.4 Qualifying a Product The 3 stearate products are ambiguous. A Qualification Method is used to determine how similar a sample is it to what has been accepted in the past. It can also be used to distinguish very similar products, e. g. Stearates. Create a Qualification Method 1.
1.5 Library Validation – Testing the Library Model Figure 3: 1 Results Qualification Method Sample Spectra 2 Histogram Distance versus Frequency Color Code: Dark blue = Training Set; samples determined to be suitable for library development. Light blue = Acceptance Set; samples suitable for library development but which are redundant compared with those in the Training set. The Acceptance set is used for Validation of the Library.
4 Qualitative Model Development Save Results 1. Click on Select Save Results 2. If asked “Do you need to save Developed Qualify Method?” click on [YES] 3. In the Library Method dialog box, click on [OK] Repeat Qualification 1. Repeat the Qualification steps for “Magnesium Stearate” and “Stearic Acid”. 2. Save the Results. 1.5 Library Validation – Testing the Library Model Library validation is used to test the library model. Open the Library 1.
1.5 Library Validation – Testing the Library Model Save Results 1. In the Library Validation dialog box, click on [Save]. 2. The software will tell you “Validation results saved in library”. Click on [OK] then click on [Close].
5 Quantitative Model Development 5 Quantitative Model Development Quantitative Analysis mode is the part of Vision where calibration equations are developed and tested. The created calibration equations can be used in Routine Analysis to predict unknown sample composition in real time. Vision offers two types of calibration development: Multilinear Regression (MLR) and Partial Least Squares (PLS). 1.
1.6 Multilinear Regression Perform Sample Selection 1. Click on Select Apply Math. 2. In the Edit Sample Selection Parameters dialog box enter the following settings: Selection Method Mahalanobis Distance in Principal Component Space Math Treatment 2nd derivative Leave all other settings unchanged 3. Click on [OK]. 4.
5 Quantitative Model Development Figure 5 Result Sample Selection for quantitative Analysis 1 Math pre-treated spectra 2 Explained Variance 3 Sample location in 3-D Principal Component Space 4 Frequency versus Probability Color Code: Dark blue = Training Set; Samples suitable for calibration development. Light blue = Acceptance Set; Samples suitable for calibration but redundant. Red = Rejected; Samples unsuitable for calibration development.
1.6 Multilinear Regression Save Results 1.6.2 1. Click on Select Save Results. 2. If you are asked “Warning – By saving this calibration equation you will invalidate the regression method which you previously developed for this equation. You must develop the regression again to use this equation. Do you want to continue?”, click on [YES]. 3. If you are asked “Do you want to save redundant samples as Validation Set?” Click on [YES]. Regression Configure Regression Method 1.
5 Quantitative Model Development Save Equation 1. Click on Regression Save Equation 2.
1.7 Partial Least Squares Regression Correlation: Generally, the selected wavelength should be where the absolute value of correlation is high. When using 2nd derivative spectra, locate an area of high negative correlation. (When using raw spectra, use a wavelength with high positive correlation.) The wavelength should be selected in a region where correlation is relatively broad and flat, not narrow and spike-like.
5 Quantitative Model Development Select a Product 1. Click on Mode Quantitative Analysis Sample Selection. 2. Double click on the “xdsstart” project 3. Click and drag the product “HiPro” onto the red EQ and release the mouse button. 4. In the Select Appropriate Constituent dialog constituent Protein is listed. Enter: Unit % 5. Click on [OK] 6. In the Calibration Equation dialog box, click on [OK] Notice Protein now appears under the red Eq and the cuvette is gray.
1.7 Partial Least Squares Regression Training Set Acceptance Set Click on [OK] 75 % 25 % Vision now performs calculations according to the parameters defined in the previous steps. The Results of the Sample Selection are displayed (cf. Figure 5, page 20).Figure 5 Result Sample Selection for quantitative Analysis Save Results 1.7.2 1. Click on Select Save Results. 2.
5 Quantitative Model Development 8. Click on [OK]. 9. In the Regression Treatment Properties dialog box, in the Wavelength Regions tab, enter the following settings: Region 1 Minimum: 750 Maximum: 1050 Region 2 Minimum:1150 Maximum: 1350 Region 3 Minimum: 1500 Maximum: 1830 Region 4 Minimum: 1990 Maximum: 2450 10. In the Edit Validation tab Method Cross Validation Remove only one sample at a time enabled Click on [OK].
1.7 Partial Least Squares Regression 1 Loading weights and coefficients 2 Statistical results 3 Statistical results with statistical summary 4 PRESS and SEC The program automatically selects the number of factors for the method based on the approach of the PRESS (Predicted Residual Error Sum of Squares) value to a minimum.
6 Routine Analysis 6 Routine Analysis In Routine Analysis you use the models and equations you have built to analyze unknown samples. All Routine Analysis parameters are brought together in the Operations Method. 1.8 Setting up an Operations Method Vision offers a wizard for setting up an operations Method. Start Operations Method Wizard 1. Click on Mode Data Acquisition 2. Click on Mode Routine Analysis OM Wizard. 3.
1.8 Setting up an Operations Method 3. In the OM Wizard – Select DCM dialog box, click on [Next]. Define an Output Project 1. In the OM Wizard – Output Project dialog box, click on [Output Project] 2. In the Choose Output Project dialog box, click on [New Output Project] 3. In the Create New Output Project dialog box, enter Project ID “Testout” Click on [OK] If asked „Create database C:\vision\testout?“, click on [Yes] 4.
6 Routine Analysis Quantification Setup 1. In the OM Wizard – Quant Setup dialog box, click on [Quant Setup] 2. In the Operations Method Product Selection dialog box, do the following: 3. Repeat for Product HiPro: 4.
1.9 Analyzing Stored Data 1.9 Analyzing Stored Data In the Analyze Stored Data mode, it is not necessary to have an instrument connected. Vision will apply your previously configured Operations Method to the samples selected and predict the constituents. The results are displayed in the lower window. Results for the library will report in the ID Result and Qual Result.
6 Routine Analysis Exit Routine Analysis/Analyze Stored Data Mode 1. Click on Mode Data Acquisition 1.10 Recalling Stored Results The Recall Results mode permits retrieval of Routine analysis results, provided the operations method specifies that results are to be saved. Samples stored in this fashion are identifiable by time and date. 1. Click on Mode Routine Analysis Recall Results. 2.
7.1 21 CFR Compliant Mode 7 Appendix 7.1 21 CFR Compliant Mode If you have chosen to operate in the 21CFR Compliant Mode, your Account Policy information should be entered at this time. Go to the Vision toolbar and select Configure followed by Account Policy and access the Account Policy setup screen. Your company establishes these policies.
