Cargando…
HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis
BACKGROUND: The vascular disorder Hereditary Hemorrhagic Telangiectasia (HHT) is in general an inherited disease caused by mutations in the TGF-β/BMP receptors endoglin or ALK1 or in rare cases by mutations of the TGF-β signal transducer protein Smad4 leading to the combined syndrome of juvenile pol...
Autores principales: | , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
BioMed Central
2013
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3698163/ https://www.ncbi.nlm.nih.gov/pubmed/23805858 http://dx.doi.org/10.1186/1750-1172-8-94 |
_version_ | 1782275255982620672 |
---|---|
author | Lux, Andreas Müller, Ralf Tulk, Mark Olivieri, Carla Zarrabeita, Roberto Salonikios, Theresia Wirnitzer, Bernhard |
author_facet | Lux, Andreas Müller, Ralf Tulk, Mark Olivieri, Carla Zarrabeita, Roberto Salonikios, Theresia Wirnitzer, Bernhard |
author_sort | Lux, Andreas |
collection | PubMed |
description | BACKGROUND: The vascular disorder Hereditary Hemorrhagic Telangiectasia (HHT) is in general an inherited disease caused by mutations in the TGF-β/BMP receptors endoglin or ALK1 or in rare cases by mutations of the TGF-β signal transducer protein Smad4 leading to the combined syndrome of juvenile polyposis and HHT. HHT is characterized by several clinical symptoms like spontaneous and recurrent epistaxis, multiple telangiectases at sites like lips, oral cavity, fingers, nose, and visceral lesions like gastrointestinal telangiectasia, pulmonary, hepatic, cerebral or spinal arteriovenous malformations. The disease shows an inter- and intra-family variability in penetrance as well as symptoms from mild to life threatening. Penetrance is also depending on age. Diagnosis of the disease is based on the presence of some of the listed symptoms or by genetic testing. HHT diagnosis is laborious, time consuming, costly and sometimes uncertain. Not all typical symptoms may be present, especially at a younger age, and genetic testing does not always identify the disease causing mutation. METHODS: Infrared (IR) spectroscopy was investigated as a potential alternative to the current diagnostic methods. IR-spectra were obtained by Fourier-transform Mid-IR spectroscopy from blood plasma from HHT patients and a healthy control group. Spectral data were mathematically processed and subsequently classified and analysed by artificial neural network (ANN) analyses and by visual analysis of scatter plots of the dominant principal components. RESULTS: The analyses showed that for HHT a disease specific IR-spectrum exists that is significantly different from the control group. Furthermore, at the current stage with the here used methods, HHT can be diagnosed by Mid-IR-spectroscopy in combination with ANN analysis with a sensitivity and specificity of at least 95%. Visual analysis of PCA scatter plots revealed an inter class variation of the HHT group. CONCLUSION: IR-spectroscopy in combination with ANN analysis can be considered to be a serious alternative diagnostic method compared to clinical and genetically based methods. Blood plasma is an ideal candidate for diagnostic purposes, it is inexpensive, easy to isolate and only minimal amounts are required. In addition, IR-spectroscopy measurement times are fast, less than one minute, and diagnosis is not based on interpretation of may be uncertain clinical data. And last but not least, the method is inexpensive. |
format | Online Article Text |
id | pubmed-3698163 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2013 |
publisher | BioMed Central |
record_format | MEDLINE/PubMed |
spelling | pubmed-36981632013-07-02 HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis Lux, Andreas Müller, Ralf Tulk, Mark Olivieri, Carla Zarrabeita, Roberto Salonikios, Theresia Wirnitzer, Bernhard Orphanet J Rare Dis Research BACKGROUND: The vascular disorder Hereditary Hemorrhagic Telangiectasia (HHT) is in general an inherited disease caused by mutations in the TGF-β/BMP receptors endoglin or ALK1 or in rare cases by mutations of the TGF-β signal transducer protein Smad4 leading to the combined syndrome of juvenile polyposis and HHT. HHT is characterized by several clinical symptoms like spontaneous and recurrent epistaxis, multiple telangiectases at sites like lips, oral cavity, fingers, nose, and visceral lesions like gastrointestinal telangiectasia, pulmonary, hepatic, cerebral or spinal arteriovenous malformations. The disease shows an inter- and intra-family variability in penetrance as well as symptoms from mild to life threatening. Penetrance is also depending on age. Diagnosis of the disease is based on the presence of some of the listed symptoms or by genetic testing. HHT diagnosis is laborious, time consuming, costly and sometimes uncertain. Not all typical symptoms may be present, especially at a younger age, and genetic testing does not always identify the disease causing mutation. METHODS: Infrared (IR) spectroscopy was investigated as a potential alternative to the current diagnostic methods. IR-spectra were obtained by Fourier-transform Mid-IR spectroscopy from blood plasma from HHT patients and a healthy control group. Spectral data were mathematically processed and subsequently classified and analysed by artificial neural network (ANN) analyses and by visual analysis of scatter plots of the dominant principal components. RESULTS: The analyses showed that for HHT a disease specific IR-spectrum exists that is significantly different from the control group. Furthermore, at the current stage with the here used methods, HHT can be diagnosed by Mid-IR-spectroscopy in combination with ANN analysis with a sensitivity and specificity of at least 95%. Visual analysis of PCA scatter plots revealed an inter class variation of the HHT group. CONCLUSION: IR-spectroscopy in combination with ANN analysis can be considered to be a serious alternative diagnostic method compared to clinical and genetically based methods. Blood plasma is an ideal candidate for diagnostic purposes, it is inexpensive, easy to isolate and only minimal amounts are required. In addition, IR-spectroscopy measurement times are fast, less than one minute, and diagnosis is not based on interpretation of may be uncertain clinical data. And last but not least, the method is inexpensive. BioMed Central 2013-06-27 /pmc/articles/PMC3698163/ /pubmed/23805858 http://dx.doi.org/10.1186/1750-1172-8-94 Text en Copyright © 2013 Lux et al.; licensee BioMed Central Ltd. http://creativecommons.org/licenses/by/2.0 This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. |
spellingShingle | Research Lux, Andreas Müller, Ralf Tulk, Mark Olivieri, Carla Zarrabeita, Roberto Salonikios, Theresia Wirnitzer, Bernhard HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis |
title | HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis |
title_full | HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis |
title_fullStr | HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis |
title_full_unstemmed | HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis |
title_short | HHT diagnosis by Mid-infrared spectroscopy and artificial neural network analysis |
title_sort | hht diagnosis by mid-infrared spectroscopy and artificial neural network analysis |
topic | Research |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3698163/ https://www.ncbi.nlm.nih.gov/pubmed/23805858 http://dx.doi.org/10.1186/1750-1172-8-94 |
work_keys_str_mv | AT luxandreas hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis AT mullerralf hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis AT tulkmark hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis AT oliviericarla hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis AT zarrabeitaroberto hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis AT salonikiostheresia hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis AT wirnitzerbernhard hhtdiagnosisbymidinfraredspectroscopyandartificialneuralnetworkanalysis |