Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium
Authors
ZHU, Xi, Yoojean KIM, Orren RAVID, Xiaofu HE, Benjamin SUAREZ-JIMENEZ, Sigal ZILCHA-MANO, Amit LAZAROV, Seonjoo LEE, Chadi G ABDALLAH, Michael ANGSTADT, Christopher L AVERILL, C Lexi BAIRD, Lee A BAUGH, Jennifer U BLACKFORD, Jessica BOMYEA, Steven E BRUCE, Richard A BRYANT, Zhihong CAO, Kyle CHOI, Josh CISLER, Andrew S COTTON, Judith K DANIELS, Nicholas D DAVENPORT, Richard J DAVIDSON, Michael D DEBELLIS, Emily L DENNIS, Maria DENSMORE, Terri DEROON-CASSINI, Seth G DISNER, El Hage WISSAM, Amit ETKIN, Negar FANI, Kelene A FERCHO, Jacklynn FITZGERALD, Gina L FORSTER, Jessie L FRIJLING, Elbert GEUZE, Atilla GONENC, Evan M GORDON, Staci GRUBER, Daniel GRUPE, Jeffrey P GUENETTE, Courtney C HASWELL, Ryan J HERRINGA, Julia HERZOG, David Bernd HOFMANN, Bobak HOSSEINI, Anna R HUDSON, Ashley A HUGGINS, Jonathan C IPSER, Neda JAHANSHAD, Meilin JIA-RICHARDS, Tanja JOVANOVIC, Milissa L KAUFMAN, Mitzy KENNIS, Anthony KING, Philipp KINZEL, Saskia B J KOCH, Inga K KOERTE, Sheri M KOOPOWITZ, Mayuresh S KORGAONKAR, John H KRYSTAL, Ruth LANIUS, Christine L LARSON, Lauren A M LEBOIS, Gen LI, Israel LIBERZON, Guang Ming LU, Yifeng LUO, Vincent A MAGNOTTA, Antje MANTHEY, Adi MARON-KATZ, Geoffery MAY, Katie MCLAUGHLIN, Sven C MUELLER, Laura NAWIJN, Steven M NELSON, Richard W J NEUFELD, Jack B NITSCHKE, Erin M LEARY, Bunmi O OLATUNJI, Miranda OLFF, Matthew PEVERILL, K Luan PHAN, Rongfeng QI, Yann QUIDE, Ivan REKTOR (203 Czech Republic, belonging to the institution), Kerry RESSLER, Pavel ŘÍHA (203 Czech Republic, belonging to the institution), Marisa ROSS, Isabelle M ROSSO, Lauren E SALMINEN, Kelly SAMBROOK, Christian SCHMAHL, Martha E SHENTON, Margaret SHERIDAN, Chiahao SHIH, Maurizio SICORELLO, Anika SIERK, Alan N SIMMONS, Raluca M SIMONS, Jeffrey S SIMONS, Scott R SPONHEIM, Murray B STEIN, Dan J STEIN, Jennifer S STEVENS, Thomas STRAUBE, Delin SUN, Jean THEBERGE, Paul M THOMPSON, Sophia I THOMOPOULOS, Nic J A VAN DER WEE, Steven J A VAN DER WERFF, Theo G M VAN ERP, Sanne J H VAN ROOIJ, van Zuiden MIRJAM, Tim VARKEVISSER, Dick J VELTMAN, Robert R J M VERMEIREN, Henrik WALTER, Li WANG, Xin WANG, Carissa WEIS, Sherry WINTERNITZ, Hong XIE, Ye ZHU, Melanie WALL, Yuval NERIA and Rajendra A MOREY (guarantor)
Edition
Neuroimage, San Diego, ACADEMIC PRESS INC ELSEVIER SCIENCE, 2023, 1053-8119
Background: Recent advances in data-driven computational approaches have been helpful in devising tools to objectively diagnose psychiatric disorders. However, current machine learning studies limited to small homogeneous samples, different methodologies, and different imaging collection protocols, limit the ability to directly compare and generalize their results. Here we aimed to classify individuals with PTSD versus controls and assess the generalizability using a large heterogeneous brain datasets from the ENIGMA-PGC PTSD Working group. Methods: We analyzed brain MRI data from 3,477 structural-MRI; 2,495 resting state-fMRI; and 1,952 diffusion-MRI. First, we identified the brain features that best distinguish individuals with PTSD from controls using traditional machine learning methods. Second, we assessed the utility of the denoising variational autoencoder (DVAE) and evaluated its classification performance. Third, we assessed the generalizability and reproducibility of both models using leave-one-site-out cross-validation procedure for each modality. Results: We found lower performance in classifying PTSD vs. controls with data from over 20 sites (60 % test AUC for s-MRI, 59 % for rs-fMRI and 56 % for D-MRI), as compared to other studies run on single-site data. The performance increased when classifying PTSD from HC without trauma history in each modality (75 % AUC). The classification performance remained intact when applying the DVAE framework, which reduced the number of features. Finally, we found that the DVAE framework achieved better generalization to unseen datasets compared with the traditional machine learning frameworks, albeit performance was slightly above chance. Conclusion: These results have the potential to provide a baseline classification performance for PTSD when using large scale neuroimaging datasets. Our findings show that the control group used can heavily affect classification performance. The DVAE framework provided better generalizability for the multi-site data. This may be more significant in clinical practice since the neuroimaging-based diagnostic DVAE classification models are much less site-specific, rendering them more generalizable.
ZHU, Xi, Yoojean KIM, Orren RAVID, Xiaofu HE, Benjamin SUAREZ-JIMENEZ, Sigal ZILCHA-MANO, Amit LAZAROV, Seonjoo LEE, Chadi G ABDALLAH, Michael ANGSTADT, Christopher L AVERILL, C Lexi BAIRD, Lee A BAUGH, Jennifer U BLACKFORD, Jessica BOMYEA, Steven E BRUCE, Richard A BRYANT, Zhihong CAO, Kyle CHOI, Josh CISLER, Andrew S COTTON, Judith K DANIELS, Nicholas D DAVENPORT, Richard J DAVIDSON, Michael D DEBELLIS, Emily L DENNIS, Maria DENSMORE, Terri DEROON-CASSINI, Seth G DISNER, El Hage WISSAM, Amit ETKIN, Negar FANI, Kelene A FERCHO, Jacklynn FITZGERALD, Gina L FORSTER, Jessie L FRIJLING, Elbert GEUZE, Atilla GONENC, Evan M GORDON, Staci GRUBER, Daniel GRUPE, Jeffrey P GUENETTE, Courtney C HASWELL, Ryan J HERRINGA, Julia HERZOG, David Bernd HOFMANN, Bobak HOSSEINI, Anna R HUDSON, Ashley A HUGGINS, Jonathan C IPSER, Neda JAHANSHAD, Meilin JIA-RICHARDS, Tanja JOVANOVIC, Milissa L KAUFMAN, Mitzy KENNIS, Anthony KING, Philipp KINZEL, Saskia B J KOCH, Inga K KOERTE, Sheri M KOOPOWITZ, Mayuresh S KORGAONKAR, John H KRYSTAL, Ruth LANIUS, Christine L LARSON, Lauren A M LEBOIS, Gen LI, Israel LIBERZON, Guang Ming LU, Yifeng LUO, Vincent A MAGNOTTA, Antje MANTHEY, Adi MARON-KATZ, Geoffery MAY, Katie MCLAUGHLIN, Sven C MUELLER, Laura NAWIJN, Steven M NELSON, Richard W J NEUFELD, Jack B NITSCHKE, Erin M LEARY, Bunmi O OLATUNJI, Miranda OLFF, Matthew PEVERILL, K Luan PHAN, Rongfeng QI, Yann QUIDE, Ivan REKTOR, Kerry RESSLER, Pavel ŘÍHA, Marisa ROSS, Isabelle M ROSSO, Lauren E SALMINEN, Kelly SAMBROOK, Christian SCHMAHL, Martha E SHENTON, Margaret SHERIDAN, Chiahao SHIH, Maurizio SICORELLO, Anika SIERK, Alan N SIMMONS, Raluca M SIMONS, Jeffrey S SIMONS, Scott R SPONHEIM, Murray B STEIN, Dan J STEIN, Jennifer S STEVENS, Thomas STRAUBE, Delin SUN, Jean THEBERGE, Paul M THOMPSON, Sophia I THOMOPOULOS, Nic J A VAN DER WEE, Steven J A VAN DER WERFF, Theo G M VAN ERP, Sanne J H VAN ROOIJ, van Zuiden MIRJAM, Tim VARKEVISSER, Dick J VELTMAN, Robert R J M VERMEIREN, Henrik WALTER, Li WANG, Xin WANG, Carissa WEIS, Sherry WINTERNITZ, Hong XIE, Ye ZHU, Melanie WALL, Yuval NERIA and Rajendra A MOREY. Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium. Neuroimage. San Diego: ACADEMIC PRESS INC ELSEVIER SCIENCE, 2023, vol. 283, December 2023, p. 1-13. ISSN 1053-8119. Available from: https://dx.doi.org/10.1016/j.neuroimage.2023.120412.
@article{2370300, author = {Zhu, Xi and Kim, Yoojean and Ravid, Orren and He, Xiaofu and SuarezandJimenez, Benjamin and ZilchaandMano, Sigal and Lazarov, Amit and Lee, Seonjoo and Abdallah, Chadi G and Angstadt, Michael and Averill, Christopher L and Baird, C Lexi and Baugh, Lee A and Blackford, Jennifer U and Bomyea, Jessica and Bruce, Steven E and Bryant, Richard A and Cao, Zhihong and Choi, Kyle and Cisler, Josh and Cotton, Andrew S and Daniels, Judith K and Davenport, Nicholas D and Davidson, Richard J and Debellis, Michael D and Dennis, Emily L and Densmore, Maria and deRoonandCassini, Terri and Disner, Seth G and Wissam, El Hage and Etkin, Amit and Fani, Negar and Fercho, Kelene A and Fitzgerald, Jacklynn and Forster, Gina L and Frijling, Jessie L and Geuze, Elbert and Gonenc, Atilla and Gordon, Evan M and Gruber, Staci and Grupe, Daniel and Guenette, Jeffrey P and Haswell, Courtney C and Herringa, Ryan J and Herzog, Julia and Hofmann, David Bernd and Hosseini, Bobak and Hudson, Anna R and Huggins, Ashley A and Ipser, Jonathan C and Jahanshad, Neda and JiaandRichards, Meilin and Jovanovic, Tanja and Kaufman, Milissa L and Kennis, Mitzy and King, Anthony and Kinzel, Philipp and Koch, Saskia B J and Koerte, Inga K and Koopowitz, Sheri M and Korgaonkar, Mayuresh S and Krystal, John H and Lanius, Ruth and Larson, Christine L and Lebois, Lauren A M and Li, Gen and Liberzon, Israel and Lu, Guang Ming and Luo, Yifeng and Magnotta, Vincent A and Manthey, Antje and MaronandKatz, Adi and May, Geoffery and Mclaughlin, Katie and Mueller, Sven C and Nawijn, Laura and Nelson, Steven M and Neufeld, Richard W J and Nitschke, Jack B and Leary, Erin M and Olatunji, Bunmi O and Olff, Miranda and Peverill, Matthew and Phan, K Luan and Qi, Rongfeng and Quide, Yann and Rektor, Ivan and Ressler, Kerry and Říha, Pavel and Ross, Marisa and Rosso, Isabelle M and Salminen, Lauren E and Sambrook, Kelly and Schmahl, Christian and Shenton, Martha E and Sheridan, Margaret and Shih, Chiahao and Sicorello, Maurizio and Sierk, Anika and Simmons, Alan N and Simons, Raluca M and Simons, Jeffrey S and Sponheim, Scott R and Stein, Murray B and Stein, Dan J and Stevens, Jennifer S and Straube, Thomas and Sun, Delin and Theberge, Jean and Thompson, Paul M and Thomopoulos, Sophia I and van der Wee, Nic J A and van der Werff, Steven J A and van Erp, Theo G M and van Rooij, Sanne J H and Mirjam, van Zuiden and Varkevisser, Tim and Veltman, Dick J and Vermeiren, Robert R J M and Walter, Henrik and Wang, Li and Wang, Xin and Weis, Carissa and Winternitz, Sherry and Xie, Hong and Zhu, Ye and Wall, Melanie and Neria, Yuval and Morey, Rajendra A}, article_location = {San Diego}, article_number = {December 2023}, doi = {http://dx.doi.org/10.1016/j.neuroimage.2023.120412}, keywords = {Posttraumatic stress disorder; Multimodal MRI; Machine learning; Deep learning; Classification}, language = {eng}, issn = {1053-8119}, journal = {Neuroimage}, title = {Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium}, url = {https://www.sciencedirect.com/science/article/pii/S1053811923005633?via%3Dihub}, volume = {283}, year = {2023} }
TY - JOUR ID - 2370300 AU - Zhu, Xi - Kim, Yoojean - Ravid, Orren - He, Xiaofu - Suarez-Jimenez, Benjamin - Zilcha-Mano, Sigal - Lazarov, Amit - Lee, Seonjoo - Abdallah, Chadi G - Angstadt, Michael - Averill, Christopher L - Baird, C Lexi - Baugh, Lee A - Blackford, Jennifer U - Bomyea, Jessica - Bruce, Steven E - Bryant, Richard A - Cao, Zhihong - Choi, Kyle - Cisler, Josh - Cotton, Andrew S - Daniels, Judith K - Davenport, Nicholas D - Davidson, Richard J - Debellis, Michael D - Dennis, Emily L - Densmore, Maria - deRoon-Cassini, Terri - Disner, Seth G - Wissam, El Hage - Etkin, Amit - Fani, Negar - Fercho, Kelene A - Fitzgerald, Jacklynn - Forster, Gina L - Frijling, Jessie L - Geuze, Elbert - Gonenc, Atilla - Gordon, Evan M - Gruber, Staci - Grupe, Daniel - Guenette, Jeffrey P - Haswell, Courtney C - Herringa, Ryan J - Herzog, Julia - Hofmann, David Bernd - Hosseini, Bobak - Hudson, Anna R - Huggins, Ashley A - Ipser, Jonathan C - Jahanshad, Neda - Jia-Richards, Meilin - Jovanovic, Tanja - Kaufman, Milissa L - Kennis, Mitzy - King, Anthony - Kinzel, Philipp - Koch, Saskia B J - Koerte, Inga K - Koopowitz, Sheri M - Korgaonkar, Mayuresh S - Krystal, John H - Lanius, Ruth - Larson, Christine L - Lebois, Lauren A M - Li, Gen - Liberzon, Israel - Lu, Guang Ming - Luo, Yifeng - Magnotta, Vincent A - Manthey, Antje - Maron-Katz, Adi - May, Geoffery - Mclaughlin, Katie - Mueller, Sven C - Nawijn, Laura - Nelson, Steven M - Neufeld, Richard W J - Nitschke, Jack B - Leary, Erin M - Olatunji, Bunmi O - Olff, Miranda - Peverill, Matthew - Phan, K Luan - Qi, Rongfeng - Quide, Yann - Rektor, Ivan - Ressler, Kerry - Říha, Pavel - Ross, Marisa - Rosso, Isabelle M - Salminen, Lauren E - Sambrook, Kelly - Schmahl, Christian - Shenton, Martha E - Sheridan, Margaret - Shih, Chiahao - Sicorello, Maurizio - Sierk, Anika - Simmons, Alan N - Simons, Raluca M - Simons, Jeffrey S - Sponheim, Scott R - Stein, Murray B - Stein, Dan J - Stevens, Jennifer S - Straube, Thomas - Sun, Delin - Theberge, Jean - Thompson, Paul M - Thomopoulos, Sophia I - van der Wee, Nic J A - van der Werff, Steven J A - van Erp, Theo G M - van Rooij, Sanne J H - Mirjam, van Zuiden - Varkevisser, Tim - Veltman, Dick J - Vermeiren, Robert R J M - Walter, Henrik - Wang, Li - Wang, Xin - Weis, Carissa - Winternitz, Sherry - Xie, Hong - Zhu, Ye - Wall, Melanie - Neria, Yuval - Morey, Rajendra A PY - 2023 TI - Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium JF - Neuroimage VL - 283 IS - December 2023 SP - 1-13 EP - 1-13 PB - ACADEMIC PRESS INC ELSEVIER SCIENCE SN - 10538119 KW - Posttraumatic stress disorder KW - Multimodal MRI KW - Machine learning KW - Deep learning KW - Classification UR - https://www.sciencedirect.com/science/article/pii/S1053811923005633?via%3Dihub N2 - Background: Recent advances in data-driven computational approaches have been helpful in devising tools to objectively diagnose psychiatric disorders. However, current machine learning studies limited to small homogeneous samples, different methodologies, and different imaging collection protocols, limit the ability to directly compare and generalize their results. Here we aimed to classify individuals with PTSD versus controls and assess the generalizability using a large heterogeneous brain datasets from the ENIGMA-PGC PTSD Working group. Methods: We analyzed brain MRI data from 3,477 structural-MRI; 2,495 resting state-fMRI; and 1,952 diffusion-MRI. First, we identified the brain features that best distinguish individuals with PTSD from controls using traditional machine learning methods. Second, we assessed the utility of the denoising variational autoencoder (DVAE) and evaluated its classification performance. Third, we assessed the generalizability and reproducibility of both models using leave-one-site-out cross-validation procedure for each modality. Results: We found lower performance in classifying PTSD vs. controls with data from over 20 sites (60 % test AUC for s-MRI, 59 % for rs-fMRI and 56 % for D-MRI), as compared to other studies run on single-site data. The performance increased when classifying PTSD from HC without trauma history in each modality (75 % AUC). The classification performance remained intact when applying the DVAE framework, which reduced the number of features. Finally, we found that the DVAE framework achieved better generalization to unseen datasets compared with the traditional machine learning frameworks, albeit performance was slightly above chance. Conclusion: These results have the potential to provide a baseline classification performance for PTSD when using large scale neuroimaging datasets. Our findings show that the control group used can heavily affect classification performance. The DVAE framework provided better generalizability for the multi-site data. This may be more significant in clinical practice since the neuroimaging-based diagnostic DVAE classification models are much less site-specific, rendering them more generalizable. ER -
ZHU, Xi, Yoojean KIM, Orren RAVID, Xiaofu HE, Benjamin SUAREZ-JIMENEZ, Sigal ZILCHA-MANO, Amit LAZAROV, Seonjoo LEE, Chadi G ABDALLAH, Michael ANGSTADT, Christopher L AVERILL, C Lexi BAIRD, Lee A BAUGH, Jennifer U BLACKFORD, Jessica BOMYEA, Steven E BRUCE, Richard A BRYANT, Zhihong CAO, Kyle CHOI, Josh CISLER, Andrew S COTTON, Judith K DANIELS, Nicholas D DAVENPORT, Richard J DAVIDSON, Michael D DEBELLIS, Emily L DENNIS, Maria DENSMORE, Terri DEROON-CASSINI, Seth G DISNER, El Hage WISSAM, Amit ETKIN, Negar FANI, Kelene A FERCHO, Jacklynn FITZGERALD, Gina L FORSTER, Jessie L FRIJLING, Elbert GEUZE, Atilla GONENC, Evan M GORDON, Staci GRUBER, Daniel GRUPE, Jeffrey P GUENETTE, Courtney C HASWELL, Ryan J HERRINGA, Julia HERZOG, David Bernd HOFMANN, Bobak HOSSEINI, Anna R HUDSON, Ashley A HUGGINS, Jonathan C IPSER, Neda JAHANSHAD, Meilin JIA-RICHARDS, Tanja JOVANOVIC, Milissa L KAUFMAN, Mitzy KENNIS, Anthony KING, Philipp KINZEL, Saskia B J KOCH, Inga K KOERTE, Sheri M KOOPOWITZ, Mayuresh S KORGAONKAR, John H KRYSTAL, Ruth LANIUS, Christine L LARSON, Lauren A M LEBOIS, Gen LI, Israel LIBERZON, Guang Ming LU, Yifeng LUO, Vincent A MAGNOTTA, Antje MANTHEY, Adi MARON-KATZ, Geoffery MAY, Katie MCLAUGHLIN, Sven C MUELLER, Laura NAWIJN, Steven M NELSON, Richard W J NEUFELD, Jack B NITSCHKE, Erin M LEARY, Bunmi O OLATUNJI, Miranda OLFF, Matthew PEVERILL, K Luan PHAN, Rongfeng QI, Yann QUIDE, Ivan REKTOR, Kerry RESSLER, Pavel ŘÍHA, Marisa ROSS, Isabelle M ROSSO, Lauren E SALMINEN, Kelly SAMBROOK, Christian SCHMAHL, Martha E SHENTON, Margaret SHERIDAN, Chiahao SHIH, Maurizio SICORELLO, Anika SIERK, Alan N SIMMONS, Raluca M SIMONS, Jeffrey S SIMONS, Scott R SPONHEIM, Murray B STEIN, Dan J STEIN, Jennifer S STEVENS, Thomas STRAUBE, Delin SUN, Jean THEBERGE, Paul M THOMPSON, Sophia I THOMOPOULOS, Nic J A VAN DER WEE, Steven J A VAN DER WERFF, Theo G M VAN ERP, Sanne J H VAN ROOIJ, van Zuiden MIRJAM, Tim VARKEVISSER, Dick J VELTMAN, Robert R J M VERMEIREN, Henrik WALTER, Li WANG, Xin WANG, Carissa WEIS, Sherry WINTERNITZ, Hong XIE, Ye ZHU, Melanie WALL, Yuval NERIA and Rajendra A MOREY. Neuroimaging-based classification of PTSD using data-driven computational approaches: A multisite big data study from the ENIGMA-PGC PTSD consortium. \textit{Neuroimage}. San Diego: ACADEMIC PRESS INC ELSEVIER SCIENCE, 2023, vol.~283, December 2023, p.~1-13. ISSN~1053-8119. Available from: https://dx.doi.org/10.1016/j.neuroimage.2023.120412.