Discrimination of Cancerous and Healthy Colon Tissues: A
New Laser-Based Method
Yasemin G€undogdu,1H€usn€u Alptekin,2Pınar Karabaglı,3MustafaSSahin,2and HamdiSS. Kilic1,4,5
1Faculty of Science, Department of Physics, University of Selc¸uk, Selc¸uklu, Konya 42031, Turkey 2Medical Faculty, Department of General Surgery, University of Selc¸uk, Selc¸uklu, Konya 42031, Turkey 3Medical Faculty, Department of Pathology, University of Selc¸uk, Selc¸uklu, Konya 42031, Turkey
4Directorate of High Technology Research and Application Center, University of Selc¸uk, Selc¸uklu, Konya 42031, Turkey 5Directorate of Laser Induced Proton Therapy Application and Research Center, University of Selc¸uk, Selc¸uklu, Konya
42031, Turkey
Background: Femtosecond (fs) Laser Ionisation Mass Spectrometry (fs-LIMS) on colon tissues are described and investigated using ionization/fragmentation processes in details to present a new application in this study. Linear Time of Flight (L-TOF) mass analyzer was utilized to investigate paraffin-embedded human tissue in this study. The effect of fs laser intensity on the spectral character-istics was investigated and interpreted due to mass spectra obtained using 800 nm wavelength with 90 fs pulses at 1 kHz repetition rate.
Objectives: Mass spectra of tissues were recorded from L-TOF system and then analyzed by performing a statistical approach called Principal Component Anal-ysis (PCA). The fs-LIMS method applied is proposed as a new and pioneering technology to analyze tissues using L-TOF system, as a human free fast and reliable intra-operative cancer diagnosis method for guiding surgeon to clean the edges of cancerous tissues to be applied during the surgical operation, for pathological exami-nations. Fs-LIMS provides some unique diagnosis opportunities to investigate biochemical characteristics of cancerous tissues leading to obtain sensitive, fast, and reliable results. The analysis of tissue is based on distribution of molecular ion (m/z) peaks in low mass region (<m/z 100) in mass spectra. Fs-LIMS provides a great data for identification of tissues (healthy and cancerous) in details. The effect of laser pulse was investigated in this study and also different types of lasers are utilized for various investigations from surgery to spectroscopy.
Methods: The experimental setup mainly consists of an ultrafast (90 fs) laser system, a mass spectrometer, laser-tissue interaction/ablation chamber, data collection, and analysis system with windows based fast digital-storage oscilloscope, statistical application codes which have been developed by our group for analyses running under MATLAB software.
Results: Paraffin embedded colon tissues have been distinguished from each other with statistical approaches using fs-LIMS based data as an alternative to histological and pathological examinations.
Conclusion: The fs-LIMS method provides a powerful novel approach to identify and analyze tissues. This promising method provides a fast and reliable (free of human mistakes) diagnosis and guidance for pathologists, surgeons, and patients during surgical operations, as well as increase the significance of mass spectrometric tissue analysis methods, especially with those capability of molecular identification of tissues. Lasers Surg. Med. © 2018 Wiley Periodicals, Inc.
Key words: fs-LIMS; colon tissue; PCA; tumor; non-tumor
INTRODUCTION
Despite the fact that the incidence of human cancers is increasing worldwide, mortality from cancer has generally been reducing over the last few decades due to the technological developments in advanced diagnosis and therapy methods. According to the World Health Organi-zation (WHO) estimated rate of the incidence from colorectal cancers both of male and female in worldwide is about 9.7% and also the estimated rate of mortality from colorectal cancers is about 8.5% in 2012 statistics [1]. According to the cancer researches carried out in the last few years after 2012, it can be clearly noticed that despite the increase in the number of cases, the mortality rate decreases. This decrease is likely to continue inversely with worldwide developments in diagnosis and treatment possibilities recent years [2,3]. These rates show that
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Contract grant sponsor: Scientific Research Projects Coordina-tion Unit of Selc¸uk University; Contract grant numbers: 15201096.
Correspondence to: Hamdi S¸ . Kilic, Faculty of Science, Department of Physics, University of Selc¸uk, Selc¸uklu, Konya 42031, Turkey. E-mail: hamdisukurkilic@selcuk.edu.tr
Accepted 11 October 2018
Published online in Wiley Online Library (wileyonlinelibrary.com).
DOI 10.1002/lsm.23033
developments in advanced imaging technologies, endo-scopic, and colonoscopic investigations in the pathologic examinations of tissues are speeded up the treatment process [4–6]. Furthermore, some scientific&specific re-searches such as Rapid Evaporative Ionization Mass Spectrometry (REIMS), probe electrospray ionization mass spectrometry (PESI-MS), desorption electrospray ionization (DESI), Matrix-assisted laser desorption ioniza-tion mass spectrometry (MALDI-MS) techniques were claimed and achieved for detection of precancerous situations, but these advanced and powerful techniques require extra time to analyze tissue samples since they need sample preparation process and this process is time-consuming, it should be considered as disadvantage during operation [7–11]. The pathological examination of suspi-cious tissue samples must be analyzed as quick, reliable, and certain as possible for results to give effective decision during surgical operation. Some fresh tissues of muscle from cow and sheep and bones have been investigated without any sample preparation process. We have applied experimental work on exactly fresh samples in a recent work [12].
The use of mass spectrometry (MS) technique dates back to beginning of 19th century. The laser ionization mass spectrometry (LIMS) technique has been widely used after the laser was discovered in 1960s [13]. LIMS was initially applied by using microsecond (ms) and nanosecond (ns) laser pulses, however, ion peaks in recorded spectra show some expanding ion intensities and due to the efficient fragmentation of the parent ion. In mass spectra obtained by using fs laser pulses, especially during fragmentation/ ionization and dissociation in the chemicals, the expansion of the ionic intensities can be achieved to a minimum without breaking the parent ion [14,15]. Laser based mass spectroscopic investigations show that pulse duration affects ion formation mechanisms. It is in accordance with gained experiences and the studies in the literature show that the peaks of the ions obtained using fs laser pulses were found to be more efficient compared to pattern in recorded mass spectra in ns or shorter laser pulse durations as a result of dissociation and fragmentation processes [14,15]. Furthermore, fs laser has been utilized to transform macromolecules into the gas form for atmospheric pressure mass spectrometry. Electrospray ionization based mass spectroscopy technique was com-bined with fs laser (800 nm, 70 fs) having intensities about 1013W/cm2which was used to analyses macromolecules, and the parent ion was dominant in spectra because of fs laser pulses (50 fs) can give an opportunity of non-thermal damage process compared to ns laser pulse durations. However, disadvantages of mentioned study, samples preparation process takes a number stage for them to be analyzed using this method [16–18].
Mass spectroscopic investigations on tissues for some characteristic studies have been carried out for a number of decades [19]. The results obtained by mass spectrometry are independent of the user; it is a very powerful and promising system. Human being studies using mass spectroscopy techniques provide important knowledge
characterizing chemical structure of tissues. MALDI-MS based investigations are capable of chemical distributions of the tissues, but the handicap of this technique is that these technique offers complex preliminary preparation and time-consuming process, so it is not appropriate in situations where rapid results need to be obtained [12,20]. Laser desorption ionization mass spectrometry (LDI-MS) are used the analysis of molecules up to m/z 2000 with different laser wavelength from UV to infrared region in laser and the technique allows to research plant tissue, bacterial cells, and human tissues [18,21–23].
In this paper, authors are aimed to show some basic differences between healthy and cancerous colon tissues by using fs-LIMS. This technique may allow surgeons to clearly distinguish borders between tumor and healthy tissues and it will reduce significantly the need for additional surgical procedures during the operation in future. A comparative evaluation of healthy and cancerous colon tissues will be discussed to present a right process without any sample preparation with some significant advantages and limitations. Fs-LIMS method used in this study have been developed bin this work and applied at the first time to investigate different types of fresh (bone, lamb, and cow) tissues supplied from the butcher and details of study have been given in a recent article [12].
The main subject of this study is to distinguish colon cancer and healthy tissues with a rapid and reliable procedure by using fs-LIMS. Therefore, our study was firstly performed with the permission of the ethics committee by using the tissues in paraffin taken from pathology department of Selcuk University Medical Faculty Hospital. The laser beam has been focused on the tissue to forms a plasma from tissue consisting of biological molecules, ions, and electrons, which are pumped to the ionization region of a mass spectrometer by a high performance turbo pump.
MATERIALS AND METHODS
The experimental setup mainly consists of an ultrafast (90 fs) pulsed laser system, a mass spectrometer designed and produced in local industry by our group, laser-tissue interaction chamber (sample ablation chamber shown in Fig. 1b), data collection and analysis system with computer based digital oscilloscope, statistical application codes which have been developed by our group for analyses running under MATLAB software. Details of our experi-mental system is given below.
Femtosecond Laser System
The fs laser system is composed of an oscillator laser system (Quantronix, Ti:Light, NY) producing 90 fs laser pulses at 800 nm wavelength with 85 MHz repetition rate and used to pump an amplifier Integra-C (Quantronix, Integra-C-3.5, NY) laser system which produces 90 fs laser pulses at 800 nm wavelength with 1–3 kHz repetition rates. After the Integra-C system, 3.5 W pulse power is available and it can be controlled using a circular neutral density filter and measured using a (Gentec UP19K-30H-W5-D0)
power meter. This fs laser system was used as an energy source to produce ions via multiphoton excitation and ionization processes.
Time of Flight Mass Spectrometer
A linear time of flight mass spectrometer (L-TOF-MS) has been used to detect ions and acquire mass spectra as a function of the time of flight or mass to charge (m/z) ratio for each ion shown in Figure 1. This L-TOF-MS consists of a sample introduction system, an ion source and a 120 cm length field-free flight tube. The positive ions of the sample produced through multiphoton dissociative ionization process is extracted from the ionization region, acceler-ated, and focused to the detector through an applied electric field on extraction and focusing electrostatic lenses. The first plate is a pusher plate and maintained at a voltage of 4–5 kV, the second and third lenses at 1 kV and 0–100 V, respectively. The details of the system have been given elsewhere [24–26].
Data Analysis
The background pressure in the vacuum chamber can be pumped down to a background pressure of several times 108mbar before sample introduction and during the experi-ment and data taken started; the pressure was kept very stable about several times (in general 4–5 times) 106mbar. In addition, all the parameters of the mass spectrometer are set to give the best mass and space resolution.
The statistical analyses and characterization of biologi-cal molecules are the most important topic to investigate its remarkable properties. Statistical methods allow us to interpret the data presenting the molecular characteristics obtained mass spectra using fs-LIMS.
In Figure 1, it is shown that the experimental setup used in L-TOF MS studies to investigate mass spectra of colon tissues. Our experimental setup consists of two sections, one of them is the ionization region of mass spectrometry, the
second region is ablation of sample section. Fs laser beam at 800 nm wavelength was used for experimental investiga-tion. Beam sampler was used to divide laser beam into two arms. One of them was focused on sample cell to ablate and evaporate sample and second beam was focused on the sample introduced into vacuum chamber at the ionization region. In our studies, all background signals were recorded before the sample was irradiated, and signals obtained from samples were recorded using laser power/intensities around 0.2 W (3.92 1014W/cm2) for sample ablation and 1.8 W (3.53 1015W/cm2) for the ionization.
Ten centimeters focal length lens was used to focus laser beam on the colon tissue sample. Mass spectra obtained from colon tissues present very stable m/z values. In this study, statistical analyses were carried out by using the low mass region peaks in mass spectra< m/z 100. All of the human tissue materials concerning about this study were approved by the ethical committee of Selcuk University. Preparation Process of Tissues
The study included 20 patients with colon adenocarci-noma who underwent resective surgery at the Hospital of medical school of Selcuk University by general surgeons. The paraffin-embedded (PE) human tissue samples which represent tumor and nontumoral region of the colon were selected by a experienced pathologist. Ten micrometer thick sections were cut from the PE blocks; and each section was transferred to a glass slide. The pathologist marked the tumor and nontumoral areas on the glass slide. Paraffin embedded tissues were achieved due to permis-sion of the ethics committee but the experimental setup shown in Figure 1 allows the sample to be analyzed at high vacuum conditions without any preliminary studies. Background of PCA Approaches
PCA, which is known as a dimension reduction method, decrease the dimensionality of data while keeping the most of variations in the data [27].
Steps of the statistical method used in this work are presented as;
fs
LIMS ! Raw Mass Spectral Data !
Data Pre proces
sing ! PCA !
Results
! Plotting
Firstly, fs-LIMS ionization process was achieved and then raw mass spectral data were recorded in experimen-tal section. Data were recorded by a four channel digiexperimen-tal- digital-storage oscilloscope (Le-Croy, WaveRunner 64Xi, 600 MHz, NY) using MATLAB program written by our group for data processing.
Fundamental equation for PCA is given as
Y ¼ WX
whereW determines matrix of coefficients calculated in the Score Coefficients. In PCA, variables are named principal
Fig. 1. Experimental setup used for determination of different colon tissues applying PCA method in (a) with femtosecond laser system used in this study (b) laser beam focusing condition on the sample to be ablated by laser pulse.
components. PCA consists of matrix notation shown in equation as X which consists of n observations (rows) on m variables (columns) that were formed using principal component (chosen m/z values). We have used ion intensi-ties to construct principal components and form nxm matrix. In equation, Y is described as the matrix of scores [28–30]. The score represents the resultant data corresponding to each principal component and finally, the results can be analyzed, plotted, and interpreted.
We have applied PCA statistics to data obtained using the program developed by our group in MATLAB. Since the program has the ability to take data from oscilloscope and analyses in real time during the experiment simulta-neously, the data acquisition process can be performed in microseconds, while the data analysis process using PCA statistic can be performed in a couple of minutes [27,31]. RESULTS AND DISCUSSION
Analysis of paraffin embedded colon tissue section is resulted in a number of steps in process. The detection of mass spectra for organic compounds was carried out using a Micro Channel Plate (MCP) detector and the signal from output of MCP detector was taken directly to the input of a four channel storage digital oscilloscope which was triggered by laser beam. The digital output signal from oscilloscope was processed by MATLAB codes written our group. For the data processing, obtained compounds were selected to apply PCA as principal components that have not been previously reported these selected ions (< m/z 100).
Figure 2 The spectra in the mass range from m/z 60 to m/ z 100, obtained from colon cancer tissue section (red) and healthy tissue section (blue) and also the background signal is recorded as a green spectrum.
The fs-LIMS spectra of healthy and cancerous colon tissues are depicted in Figure 2 which recorded in positive
mode 800 nm wavelength with 90 fs laser pulses using 3.53 1015W/cm2 laser intensity. The mass range is selected from m/z 60 to m/z 100 region to show difference between relative abundance of spectra. The background signal has no any ion intensity indicating that the background signal is very clear and no peaks in this mass region appear before sample was introduced into spectrometer. That is, the spectrum coming from the sample is observed and recorded solely. All mass spectra obtained from healthy and cancer colon tissues show similar ion distributions in this region. The ions as shown in Figure 2 have m/z 64, 65, 66, 67, 68, 69, 70, 71, 76, 77, 78, 79, 80, 81, 83, 92, and 93 values and these m/z values were used in PCA statistics as principal components.
The PCA plots given in Figure 3 shows the normal and cancerous colon tissue results which were obtained from paraffin embedded block tissues and tissues were also sectioned on microscope glass slide surface. Colon tissues obtained from paraffin embedded block samples are shown in blue (nontumoral) and turquoise (tumor) colors, and also colon tissues obtained from sectioned on glass slide is shown in red (nontumoral) and yellow (tumor) colors. These configurations of tissues are important for interpre-tation of obtained results from both tissue samples. It shows that the results obtained from both tissues are in harmony with each other. According to PC1 axes, paraffin embedded tissue and sectioned tissue on glass slide surface with the tumor as well as paraffin embedded block tissues and sectioned tissues on glass slide surface with the nontumoral tissue coincided with each other on the same side. It means that the results obtained are consistent with each other.
In this section, PCA statistical results were achieved at the 3.53 1015W/cm2 laser intensities for 1.8 W laser power and obtained experimental results were used to
Fig. 2. The spectra in the mass range from m/z 60 to m/z 100, obtained from colon cancer tissue section (red) and healthy tissue section (blue)and also the background signal is recorded as a green spectrum.
Fig. 3. Colon tissue obtained from paraffin embedded block sample is shown as blue (healthy) and turquoise (cancerous) and also colon tissue obtained from sectioned on glass slide surface is shown as red (healthy) and yellow (cancerous).
evaluate the statistical approaches. We use two axes to evaluate each group for the obtained data. Each group of data have been organized by taking 10 different experi-mental results. Each point for all isomers are shown in Figure 4 including some different mass components chosen among mass ion peaks in mass spectra.
A classification model built up using PCA statistical approaches, spectral data acquired from cancerous colon tissues surrounded by healthy tissues which were consid-ered to be healthy tissues to compare them. Mass spectra were recorded to identify their spectral differences between tissue types and 17 different ion peaks responsible for classification of healthy and cancerous colon tissues in Figure 4.
PCA analysis results are shown in Figure 4, only one human colon cancer tissue and surrounding 17 human healthy colon tissues are given in comparison. It can clearly be seen from data that results from seventeen healthy tissues are all given reproducible data and can be clearly distinguished from cancerous tissue data. PCA was performed on 180 data points in this study shown in Figure 4. According to PC2 axes, colorful area is symbolized colon healthy tissues (17 tissue samples) sectioned on glass slide and also red dots represent the
cancer tissue in Figure 4 fall higher range of screen. PCA results suggest that some colorful healthy tissues are separated form cancerous tissue sample which are taken from paraffin embedded tissue on glass slide surface. Figure 4b shows the distribution of groups due to the axes. Healthy colon tissues and only one cancerous tissue can be differentiated from each other clearly using PCA statistics. Scattering results as indicated in Figure 4b, m/z 66, 68, and 70 peaks used are the main peaks to differentiate from nontumoral tissues to tumor. As it can be seen from Figure 5a that, 10 healthy results from different patients are all comes together and are clearly differentiated from the result taken from a cancerous tissue.
The fs-LIMS data were analyzed by PCA, and the obtained results are shown in Figure 5. As it is indicated in the Figure 5, data points accumulated from paraffin embedded colon tissues and sectioned on glass slide surface. According to PC2 axes, colorful area is symbolized cancerous colon tissue sectioned on glass slide surface and also red dots represent the healthy colon tissue in Figure 5. Classification of loading values determined by using PCA displayed the important mass spectral properties for classification, as shown in Figure 5a. The presented results
Fig. 4. Normalized 17 healthy colon tissues and a cancer tissue can be differentiated from each other clearly using PCA statistical approaches, (b) Scattering of tissue ions can be distinguished according to PC1 and PC2 axes.
Fig. 5. Normalized 11 cancer tissues and a healthy colon tissue and a can be differentiated from each other clearly using (a) PCA statistical approaches, (b) Scattering of tissue ions can be distinguished according to PC1 and PC2 axes.
demonstrate that fs-LIMS investigation of colon tissues is a new approach to analyze tissue samples as an alternative to pathologic investigations. Figure 5b shows the distribu-tion of groups due to the axes. Cancerous colon tissues and only one healthy tissue can be differentiated from each other clearly using PCA statistics. Scattering results as indicated in Figure 5b, m/z 67, 68, and 77 peaks used are the main peaks to differentiate from healthy tissues to cancer tissue. As it can be seen from Figure 4a that, 10 cancerous results from different patients are all comes together and are clearly differentiated from the result taken from a healthy tissue.
CONCLUSION
The fs-LIMS method provides a powerful novel approach to analyses tissue identification and this novel method is providing fast diagnosis, which is free of human mistakes, and guidance to pathologists, surgeons and patients during the operation, as well as increase the significance of mass spectrometric tissue analysis methods, especially with those capability of molecular identification of tissues. Actually we have already tested and given a report in literature that our proposed new technique is quite successful to distinguish fresh tissues of muscle from cow and sheep and bones [12] as well as paraffin embedded cancerous and healthy tissues.
The fs-LIMS method has been optimized for analysis of cancerous and healthy colon tissues embedded on paraffin and in block, and the results give that spectral analysis is both reliable and rapid. Applied method is capable of introducing samples into mass spectrometer and collection and analysis of data from biological and chemical samples. Further studies are in plan and will be applied on tissues without paraffin and may also be carried out intra operatively on the patient due to the ethics regulations in a very short time using laser based MS.
ACKNOWLEDGMENT
Authors kindly would like to thank, Scientific Research Projects Coordination Unit of Selc¸uk University for financial support via Projects No. 15201096 and 17401174. Selc¸uk University, High Technology Research and Application Center for supplying with and University of Selc¸uk, SULTAN Center for infrastructure.
References
1. Organization, W.H. 2018 [cited 2018 23.04]; http://gco.iarc.fr/ today/online-analysis-pie?mode¼cancer&mode_population¼ who&population¼900&sex¼0&cancer¼29&type¼1& statistic¼0&prevalence¼0&color_palette¼default].
2. Arnold M, Sierra MS, Laversanne M, et al. Global patterns and trends in colorectal cancer incidence and mortality. Gut 2016;66(4):683–691.
3. Siegel RL, DeSantis C, Jemal A. Colorectal cancer statistics. CA Cancer J Clin 2017;67(3):177–193.
4. Shaukat A, Mongin SJ, Geisser MS, et al. Long-term mortality after screening for colorectal cancer. N Engl J Med 2013;369(12):1106–1114.
5. Chadwick G, Groene O, Hoare J, et al. A population-based, retrospective, cohort study of esophageal cancer missed at endoscopy. Endoscopy 2014;46(07):553–560.
6. Zorzi M, Senore C, Da Re F, et al. Quality of colonoscopy in an organised colorectal cancer screening programme with immunochemical faecal occult blood test: the EQuIPE study (Evaluating Quality Indicators of the Performance of Endos-copy). Gut 2014;64(9):1389–1396.
7. Balog J, Kumar S, Alexander J, et al. In vivo endoscopic tissue identification by rapid evaporative ionization mass spectrom-etry (REIMS). Angew. Chem 2015;54(38):11059–11062. 8. Mandal MK, Saha S, Yoshimura K, et al. Biomolecular
analysis and cancer diagnostics by negative mode probe electrospray ionization. Analyst 2013;138(6):1682–1688. 9. Takats Z, Wiseman JM, Cooks RG. Ambient mass
spectrome-try using desorption electrospray ionization (DESI): instru-mentation, mechanisms and applications in forensics, chemistry, and biology. J Mass Spectrom 2005;40(10): 1261–1275.
10. Mirnezami R, Spagou K, Vorkas PA, et al. Chemical mapping of the colorectal cancer microenvironment via MALDI imaging mass spectrometry (MALDI-MSI) reveals novel cancer-associated field effects. Mol Oncol 2014;8(1):39–49. 11. St John ER, Balog J, McKenzie JS, et al. Rapid evaporative
ionisation mass spectrometry of electrosurgical vapours for the identification of breast pathology: towards an intelligent knife for breast cancer surgery. Breast Cancer Res 2017;19(1): 59.
12. G€undogdu Y, Kılıc¸ H. A New Method for Investigation of Different Tissues Using Femtosecond Laser Mass Spectrome-try. Photomed Laser Surg 2018;36(10):548–554.
13. Maiman TH, Stimulated optical radiation in ruby. Pergamon Press LTD: Newyork, USA; 1960. p 10523.
14. Kilic HS, Ledingham KWD, Kosmidis C, et al. Multiphoton ionization and dissociation of nitromethane using femtosec-ond laser pulses at 375 and 750nm. J Phys Chem A 1997;101: 817–823.
15. Ledingham KW, Deas RM, Marshall A, et al. A comparison of femtosecond and nanosecond multiphoton ionization and dissociation for some nitro-molecules. Rapid Commun Mass Spectrom 1995;9(15):1522–1527.
16. Brady JJ, Judge EJ, Levis RJ. Mass spectrometry of intact neutral macromolecules using intense non-resonant femtosecond laser vaporization with electrospray post-ionization. Rapid Commun Mass Spectrom 2009;23(19): 151–3157.
17. Gobeli D, El-Sayed M. Change in the mechanism of laser multiphoton ionization-dissociation in benzaldehyde by changing the laser pulse width. J Phys Chem A 1985;89(15): 3426–3429.
18. Flanigan IV, PM, Radell LL, Brady JJ, et al. Differentiation of eight phenotypes and discovery of potential biomarkers for a single plant organ class using laser electrospray mass spectrometry and multivariate statistical analysis. Anal Chem 2012;84 (14):6225–6232.
19. Vogel A, Venugopalan V. Mechanisms of pulsed laser ablation of biological tissues. Chem Rev 2003;103(2):577–644. 20. Karas M, Bahr U, Gießmann U. Matrix-assisted laser
desorption ionization mass spectrometry. Mass Spectrom Rev. 1991;10(5):335–357.
21. Karas M, Hillenkamp F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons. Anal Chem 1988;60(20):2299–2301.
22. Verbueken AH, Van Grieken RE, De Broe ME, et al. Identification of inorganic and organic microliths in kidney sections by laser microprobe mass spectrometry. Analytica Chimica Acta 1987;195:97–115.
23. Walker AV, Gelb LD, Barry GE, et al. Femtosecond laser desorption ionization mass spectrometry imaging and multi-variate analysis of lipids in pancreatic tissue. Biointerphases 2018;13(3):03B416.
24. Wiley W, McLaren IH. Time-of-flight mass spectrometer with improved resolution. Rev Sci Instrum. 1955;26(12): 1150–1157.
25. G€undogdu Y, Kepceoglu A, Kilic HSS, et al. Femtosaniye Laser Pulsları Yardımıyla C nþ (n ¼#2ı3# 60) Topak _Iyonlarının Hem C 60 Hemde Grafit Hedefler Kullanılarak Ayrı Ayrı €
26. Yildirim M, Sise O, Dogan M, et al. Designing multi-field linear time-of-flight mass spectrometers with higher-order space focusing. Int J Mass Spectrom 2010;291(1):1–12. 27. Kepceoglu A, G€undogdu Y, Ledingham KWD, et al.
Identifica-tion of the isomers using principal component analysis (PCA) method. AIP Conference Proceedings. AIP Publishing; 2016. 28. Software NS, Principal Components Analysis. https://www.
ncss.com/, Universidade de Coimbra: NCSS. 23.
29. Abdi H, Williams LJ. Principal component analysis. Wiley Interdiscip Rev Comput Stat 2010;2(4):433–459.
30. Shlens J, A tutorial on principal component analysis. arXiv preprint arXiv:1404.1100, 2014.
31. Kepceoglu A, Kılıc¸ HSS. Chemical analysis program for the time of flight mass spectrometry system. Int J Appl Math Electr Comput 2016;4(Special Issue-1):57–60.
