Autism risk classification using placental chorionic surface vascular network features

Jen Mei Chang; Hui Zeng; Ruxu Han; Ya Mei Chang; Ruchit Shah; Carolyn M. Salafia; Craig Newschaffer; Richard K. Miller; Philip Katzman; Jack Moye; Margaret Fallin; Cheryl K. Walker; Lisa Croen

doi:10.1186/s12911-017-0564-8

Autism risk classification using placental chorionic surface vascular network features

Jen Mei Chang, Hui Zeng, Ruxu Han, Ya Mei Chang, Ruchit Shah, Carolyn M. Salafia, Craig Newschaffer, Richard K. Miller, Philip Katzman, Jack Moye, Margaret Fallin, Cheryl K. Walker, Lisa Croen

Bloomberg School of Public Health

Research output: Contribution to journal › Article › peer-review

10 Scopus citations

Abstract

BACKGROUND: Autism Spectrum Disorder (ASD) is one of the fastest-growing developmental disorders in the United States. It was hypothesized that variations in the placental chorionic surface vascular network (PCSVN) structure may reflect both the overall effects of genetic and environmentally regulated variations in branching morphogenesis within the conceptus and the fetus' vital organs. This paper provides sound evidences to support the study of ASD risks with PCSVN through a combination of feature-selection and classification algorithms.

METHODS: Twenty eight arterial and 8 shape-based PCSVN attributes from a high-risk ASD cohort of 89 placentas and a population-based cohort of 201 placentas were examined for ranked relevance using a modified version of the random forest algorithm, called the Boruta method. Principal component analysis (PCA) was applied to isolate principal effects of arterial growth on the fetal surface of the placenta. Linear discriminant analysis (LDA) with a 10-fold cross validation was performed to establish error statistics.

RESULTS: The Boruta method selected 15 arterial attributes as relevant, implying the difference in high and low ASD risk can be explained by the arterial features alone. The five principal features obtained through PCA, which accounted for about 88% of the data variability, indicated that PCSVNs associated with placentas of high-risk ASD pregnancies generally had fewer branch points, thicker and less tortuous arteries, better extension to the surface boundary, and smaller branch angles than their population-based counterparts.

CONCLUSION: We developed a set of methods to explain major PCSVN differences between placentas associated with high risk ASD pregnancies and those selected from the general population. The research paradigm presented can be generalized to study connections between PCSVN features and other maternal and fetal outcomes such as gestational diabetes and hypertension.

Original language	English (US)
Article number	162
Pages (from-to)	162
Number of pages	1
Journal	BMC medical informatics and decision making
Volume	17
Issue number	1
DOIs	https://doi.org/10.1186/s12911-017-0564-8
State	Published - Dec 6 2017

Keywords

Arterial network
Autism spectrum disorder risk
Boruta algorithm
Linear discriminant analysis
Placenta
Placental chorionic surface vascular network (PCSVN)
Principal component analysis
Random forest

ASJC Scopus subject areas

Health Policy
Health Informatics

Access to Document

10.1186/s12911-017-0564-8

Cite this

Chang, J. M., Zeng, H., Han, R., Chang, Y. M., Shah, R., Salafia, C. M., Newschaffer, C., Miller, R. K., Katzman, P., Moye, J., Fallin, M., Walker, C. K., & Croen, L. (2017). Autism risk classification using placental chorionic surface vascular network features. BMC medical informatics and decision making, 17(1), 162. Article 162. https://doi.org/10.1186/s12911-017-0564-8

@article{341f78fc05bf4708ac68910741715930,

title = "Autism risk classification using placental chorionic surface vascular network features",

abstract = "BACKGROUND: Autism Spectrum Disorder (ASD) is one of the fastest-growing developmental disorders in the United States. It was hypothesized that variations in the placental chorionic surface vascular network (PCSVN) structure may reflect both the overall effects of genetic and environmentally regulated variations in branching morphogenesis within the conceptus and the fetus' vital organs. This paper provides sound evidences to support the study of ASD risks with PCSVN through a combination of feature-selection and classification algorithms.METHODS: Twenty eight arterial and 8 shape-based PCSVN attributes from a high-risk ASD cohort of 89 placentas and a population-based cohort of 201 placentas were examined for ranked relevance using a modified version of the random forest algorithm, called the Boruta method. Principal component analysis (PCA) was applied to isolate principal effects of arterial growth on the fetal surface of the placenta. Linear discriminant analysis (LDA) with a 10-fold cross validation was performed to establish error statistics.RESULTS: The Boruta method selected 15 arterial attributes as relevant, implying the difference in high and low ASD risk can be explained by the arterial features alone. The five principal features obtained through PCA, which accounted for about 88% of the data variability, indicated that PCSVNs associated with placentas of high-risk ASD pregnancies generally had fewer branch points, thicker and less tortuous arteries, better extension to the surface boundary, and smaller branch angles than their population-based counterparts.CONCLUSION: We developed a set of methods to explain major PCSVN differences between placentas associated with high risk ASD pregnancies and those selected from the general population. The research paradigm presented can be generalized to study connections between PCSVN features and other maternal and fetal outcomes such as gestational diabetes and hypertension.",

keywords = "Arterial network, Autism spectrum disorder risk, Boruta algorithm, Linear discriminant analysis, Placenta, Placental chorionic surface vascular network (PCSVN), Principal component analysis, Random forest",

author = "Chang, {Jen Mei} and Hui Zeng and Ruxu Han and Chang, {Ya Mei} and Ruchit Shah and Salafia, {Carolyn M.} and Craig Newschaffer and Miller, {Richard K.} and Philip Katzman and Jack Moye and Margaret Fallin and Walker, {Cheryl K.} and Lisa Croen",

year = "2017",

month = dec,

day = "6",

doi = "10.1186/s12911-017-0564-8",

language = "English (US)",

volume = "17",

pages = "162",

journal = "BMC medical informatics and decision making",

issn = "1472-6947",

publisher = "BioMed Central",

number = "1",

}

TY - JOUR

T1 - Autism risk classification using placental chorionic surface vascular network features

AU - Chang, Jen Mei

AU - Zeng, Hui

AU - Han, Ruxu

AU - Chang, Ya Mei

AU - Shah, Ruchit

AU - Salafia, Carolyn M.

AU - Newschaffer, Craig

AU - Miller, Richard K.

AU - Katzman, Philip

AU - Moye, Jack

AU - Fallin, Margaret

AU - Walker, Cheryl K.

AU - Croen, Lisa

PY - 2017/12/6

Y1 - 2017/12/6

N2 - BACKGROUND: Autism Spectrum Disorder (ASD) is one of the fastest-growing developmental disorders in the United States. It was hypothesized that variations in the placental chorionic surface vascular network (PCSVN) structure may reflect both the overall effects of genetic and environmentally regulated variations in branching morphogenesis within the conceptus and the fetus' vital organs. This paper provides sound evidences to support the study of ASD risks with PCSVN through a combination of feature-selection and classification algorithms.METHODS: Twenty eight arterial and 8 shape-based PCSVN attributes from a high-risk ASD cohort of 89 placentas and a population-based cohort of 201 placentas were examined for ranked relevance using a modified version of the random forest algorithm, called the Boruta method. Principal component analysis (PCA) was applied to isolate principal effects of arterial growth on the fetal surface of the placenta. Linear discriminant analysis (LDA) with a 10-fold cross validation was performed to establish error statistics.RESULTS: The Boruta method selected 15 arterial attributes as relevant, implying the difference in high and low ASD risk can be explained by the arterial features alone. The five principal features obtained through PCA, which accounted for about 88% of the data variability, indicated that PCSVNs associated with placentas of high-risk ASD pregnancies generally had fewer branch points, thicker and less tortuous arteries, better extension to the surface boundary, and smaller branch angles than their population-based counterparts.CONCLUSION: We developed a set of methods to explain major PCSVN differences between placentas associated with high risk ASD pregnancies and those selected from the general population. The research paradigm presented can be generalized to study connections between PCSVN features and other maternal and fetal outcomes such as gestational diabetes and hypertension.

AB - BACKGROUND: Autism Spectrum Disorder (ASD) is one of the fastest-growing developmental disorders in the United States. It was hypothesized that variations in the placental chorionic surface vascular network (PCSVN) structure may reflect both the overall effects of genetic and environmentally regulated variations in branching morphogenesis within the conceptus and the fetus' vital organs. This paper provides sound evidences to support the study of ASD risks with PCSVN through a combination of feature-selection and classification algorithms.METHODS: Twenty eight arterial and 8 shape-based PCSVN attributes from a high-risk ASD cohort of 89 placentas and a population-based cohort of 201 placentas were examined for ranked relevance using a modified version of the random forest algorithm, called the Boruta method. Principal component analysis (PCA) was applied to isolate principal effects of arterial growth on the fetal surface of the placenta. Linear discriminant analysis (LDA) with a 10-fold cross validation was performed to establish error statistics.RESULTS: The Boruta method selected 15 arterial attributes as relevant, implying the difference in high and low ASD risk can be explained by the arterial features alone. The five principal features obtained through PCA, which accounted for about 88% of the data variability, indicated that PCSVNs associated with placentas of high-risk ASD pregnancies generally had fewer branch points, thicker and less tortuous arteries, better extension to the surface boundary, and smaller branch angles than their population-based counterparts.CONCLUSION: We developed a set of methods to explain major PCSVN differences between placentas associated with high risk ASD pregnancies and those selected from the general population. The research paradigm presented can be generalized to study connections between PCSVN features and other maternal and fetal outcomes such as gestational diabetes and hypertension.

KW - Arterial network

KW - Autism spectrum disorder risk

KW - Boruta algorithm

KW - Linear discriminant analysis

KW - Placenta

KW - Placental chorionic surface vascular network (PCSVN)

KW - Principal component analysis

KW - Random forest

UR - http://www.scopus.com/inward/record.url?scp=85045413510&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85045413510&partnerID=8YFLogxK

U2 - 10.1186/s12911-017-0564-8

DO - 10.1186/s12911-017-0564-8

M3 - Article

C2 - 29212472

AN - SCOPUS:85045413510

SN - 1472-6947

VL - 17

SP - 162

JO - BMC medical informatics and decision making

JF - BMC medical informatics and decision making

IS - 1

M1 - 162

ER -

Autism risk classification using placental chorionic surface vascular network features

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this