#Peak width fityk manual#
#Peak width fityk full#
#Peak width fityk software#

Amides I, II, and III represent the organic matrix of the bone, while the ν 1 ν 3PO 4, ν 4PO 4, and ν 2CO 3 are part of the mineral matrix. Typical FTIRM spectrum of human bone between the 450 cm − cm −1 boundary. In a bone spectrum (Figure 1), six distinct regions are identified as the vibrational response of different constituents: (i) The amide (I, II, and III) areas (1300, 1700) cm −1 corresponding to the response of the organic matrix in the bone tissue (with also a ν 3CO 3 contribution at around 1400 cm −1) (ii) The ν 1 ν 3PO 4 area (900, 1200) cm −1 corresponding to the symmetric and antisymmetric stretching of the phosphates (iii) The ν 2CO 3 area (800, 900) cm −1 corresponding to the symmetrical stretching of the carbonates (CO 3), which also contains a large HPO 4 contribution, which is removed before analysis by using a wisely placed baseline correction (iv) The ν 4PO 4 area (500, 650) cm −1 corresponding to the antisymmetric deformation of the phosphates (PO 4). Indeed, the regression may sometimes be highly sensitive and be executed multiple times for one ROI in order to find the correct fitting result.

As one can easily imagine, this technique proved to be fastidious, as correct results were typically obtained through several tries. The result of the fitting process was then left to the operator for visual verification. Previously, this fitting was done by assuming both position and characteristics of the peaks (height and FWHM), giving a starting state for the least-squares fitting method. Traditionally, the peak fitting of a bone spectrum is done by selecting a region of interest (ROI) in the bone spectrum, followed by fitting the curve with a sum of Gaussians. This is why the second derivative method is not chosen (for more explanation, cf.

#Peak width fityk full#

The second derivative method, although quite promising due to its objective and purely mathematical analysis, is too sensitive to noise (producing peaks that are not there initially) and quite inadequate in the case of peaks that are too close and with different full width at half maximum (FWHM) (the wider peak is then masked by the thinner one). We chose to stick with the peak fitting method used in. Several different methods are currently used to analyse FTIRM spectra (peak fitting method, second derivative method, etc.), and each has its advantages and limitations moreover, none are considered to be a gold standard. In addition, the spectrum analysis is laborious if done “manually.” This method combines optic, quantum mechanics, chemistry, and biology knowledge, which leads to a complex understanding that might have an impact on the conclusion of an FTIR study. FTIR microspectroscopy (FTIRM) is a promising field which characterises bone samples, as it is able to provide information on the chemical composition and structural organisation of the analysed matter without damaging it on a local level of a thin section. Introductionīone analysis methods are quite varied, ranging from the noninvasive, e.g., X-ray scanning, to the destructive, e.g., mechanical rupture tests.

The developed algorithms provide the means necessary to fully compare the results between bone FTIRM studies and between different laboratories. Results and validation proved the reliability of the automatic process.

#Peak width fityk manual#

Validation of the automatic method has been performed by comparison with the manual method. Furthermore, a rigorous framework has been established concerning the peak fitting method to obtain the systematic best potential solution.

#Peak width fityk software#

To solve this issue, software has been developed based on several algorithms to automate the analysis. However, spectrum analysis methods are often highly sensitive to small variations (e.g., boundary limits), thus implying a time-consuming and redundant analysis process. FTIR microspectroscopy (FTIRM) is a commonly used nondestructive method to characterise thin bone sections.