Automated interpretation of infrared spectra with an instrument based minicomputer

Tomellini, Sterling A.; Saperstein, David D.; Stevenson, James; Smith, Graham; Woodruff, H. B.; Seelig, Paul F.

doi:10.1021/ac00237a060

Cited by 33 publications

(13 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…IntIRpret has five main subroutines: the interferogram processing and peak selection subroutine (PUSHSUB) (8), the automated knowledge acquisition subroutine (AUTOGEN) {16), the system optimization subroutine (STO), the inference engine (PAIRS) (7,(10)(11)(12)(13)(14)(15)(16), and the final processing subroutine that subtracts spectral similarity (PAIRSPLUS) (8). Figure 1 is a flow chart of the intIRpret process, where the logic of each of the five major subroutines is diagramed.…”

Section: Resultsmentioning

confidence: 99%

“…Two limitations of PAWMI were that once a training set, consisting of a library of reference of spectra, was defined, the rules for the inference engine (PAIRS) (10)(11)(12)(13)(14)(15)(16) had to be generated manually. The second limitation was that the PAWMI compound identification software only uses peak location information.…”

mentioning

confidence: 99%

“…This paper describes a program for the identification of the principal components of mixtures based on computer assited interpretation of the mixture's infrared spectrum. This program (intIRpret), which was developed as a preliminary screening tool for unknown organics handled on hazardous waste remedial action sites, has five main subroutines: the interferogram processing and peak selection subroutine (PU-SHSUB) (8), the automated knowledge acquisition subroutine (AUTOGEN) (17), the system optimization subroutine (STO), the interpretation subroutine (PAIRS) (7,(10)(11)(12)(13)(14)(15)(16), and the final processing subroutine to subtract spectral similarity (PAIR-SPLUS) (8).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Self-training, self-optimizing expert system for interpretation of the infrared spectra of environmental mixtures

Ying¹,

Levine²,

Tomellini³

et al. 1987

Anal. Chem.

Self Cite

View full text Add to dashboard Cite

A program for the Identification of the principal components of mixtures through interpretation of the infrared mixture spectrum (lnt IRpret) was developed. Thls program, which was developed as a preiimlnary Screening tool for unknown organics handled on hazardous waste remedial action sites, has five maln subroutlnes: the interferogram processing and peak selection Subroutine (PUSHSUB), the automated knowledge acquisition subroutlne (AUTOGEN), the system optlmlzatlon subroutine (STO), the interpretation subroutine (PAIRS), and the final processing subroutine to subtract spectral simliarlty (PAIRSPLUS). Prlnclpal advantages of this system compared to those previously reported are speed, fiexlblilty, and accuracy. For a training set of 62 pure compounds and a data set of 67 four-component mixtures requiring 4154 decisions, the system correctly Identified 216 true posltlve results and 3840 true negatlves and Incorrectly identlfled 46 false posltlvies (19.4%) and 52 false negatives (1.2%).In order to satisfy the requirements of hazardous waste analysis a t Superfund and a t licensed disposal sites (I+), a program for automated waste mixture identification (PAWMI) through the interpretation of the infrared (IR) spectrum of the waste mixture was developed (7, 8) and tested on hazardous waste drum samples (9). This approach, which utilizes the speed and sensitivity of Fourier transform infrared (FT-IR) spectrometry meets many of the requirements of a near real-time, principal component screening technique for organic hazardous waste samples ( I ) .Two limitations of PAWMI were that once a training set, consisting of a library of reference of spectra, was defined, the rules for the inference engine (PAIRS) (10-16) had to be generated manually. The second limitation was that the PAWMI compound identification software only uses peak location information.An approach to the automated generation of functional group interpretation rules for PAIRS was previously developed (1 7). This system defines a value "occurrencen as the "number of peaks in a given wavenumber range divided by the number of compounds in the database containing the functionality of interest". This value was used to weight peak position information for the generation of expectation values for the presence of certain functional groups. Efforts by other investigators have been successful in the interpretation of IR spectra by using computerized inter-' Present address: . pretation or matching procedures. These program systems include the hierarchal tree (18) and table-driven (19) programs developed by Munk et al. and the pattern recognition approach of Frankel (20). These systems were primarily aimed a t identifying functional groups in compounds, as was the original PAIRS program. The success of these approaches were a function of the data set and functional group studied. A related work aimed primarily a t identifying compounds in mixtures was that of Lowry (21) which used a Boolean logic based search system. Many of these approaches owe their origins to ear...

show abstract

Section: Resultsmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Self-training, self-optimizing expert system for interpretation of the infrared spectra of environmental mixtures

Ying¹,

Levine²,

Tomellini³

et al. 1987

Anal. Chem.

Self Cite

View full text Add to dashboard Cite

show abstract

“…spectral library searching and computerized spectral interpretation as an aid to chemical structural determination, often in combination with mass spectroscopy or n.m.r. (Tomellini et al 1981;Visser & Van der Mass 1981;Craver et al 1981;Zupan et al 1980;Schaarschmidt 1979;Small et al 1979;Delany & Uden 1979;Tanabe et al 1979;Gribov et al 1979) and this is an area of continuing activity at a number of centres. Commercial i.r.…”

Section: Qualitativementioning

confidence: 99%

Computer control of equipment and data handling

Ford

1982

Phil. Trans. R. Soc. Lond. A

View full text Add to dashboard Cite

In the 10 years since the first commercially available microprocessor was announced, nearly all commercial optical and infrared spectrometers have become computer controlled. The advantages of such control and the cost of writing the necessary programs are discussed, and the industrial and academic situations contrasted. Data handling is discussed in terms of signal handling and the subsequent data-processing techniques available. Examples are given of the use of methods to aid visualization. Results are given of the application of the least-squares method to the measurement of a trace component, and an infrared interpretative search program is described.

show abstract

“…A number of programs have been developed to assist the chemist in the identification and quantitation of unknown compounds. Many approaches to automated spectral interpretation have been developed such as pattern recognition, factor analysis, hierarchical clustering, and expert systems (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18). Summaries of recent literature in this and related subjects are available in symposium proceedings (19,20).…”

Section: Introductionmentioning

confidence: 99%

Iterative least-squares fit procedures for the identification of organic vapor mixtures by Fourier-transform infrared spectrophotometry

Xiao¹,

Levine²,

D’Arcy³

1989

Anal. Chem.

View full text Add to dashboard Cite

Least-squares fitting (LSF) was applied to the qualitative analysis of IR spectra based on comparing standard reference spectra with the sample mixture spectrum. Identification of compounds in the sample was made by judging the fit level of the spectrum of each compound with the sample spectrum. An iterative procedure was developed to eliminate compounds with the worst fit levels in order to approach an optimal fit for the sample spectrum. The qualitative analysis results obtained from the optimal fit were further used for quantitative analysis.

show abstract

Automated interpretation of infrared spectra with an instrument based minicomputer

Cited by 33 publications

References 10 publications

Self-training, self-optimizing expert system for interpretation of the infrared spectra of environmental mixtures

Self-training, self-optimizing expert system for interpretation of the infrared spectra of environmental mixtures

Computer control of equipment and data handling

Iterative least-squares fit procedures for the identification of organic vapor mixtures by Fourier-transform infrared spectrophotometry

Contact Info

Product

Resources

About