Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state

Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state Alexandra Korda, Rebekka Lencer, Andreas Sprenger, Inga Meyhöfer, and 20 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7178786/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 12 Feb, 2026 Read the published version in Translational Psychiatry → Version 1 posted 9 You are reading this latest preprint version Abstract Deeper understanding of Subtle Visual Dysfunctions (VisDys) in the early stage of mental illness and their neurobiological underpinnings, as reflected by microstructural brain texture features, could advance our understanding of the underlying disease perceptual mechanisms that mediate susceptibility to psychosis. In this study, we aim a) to investigate the utility of brain texture features for the prediction of VisDys in recent onset psychosis (ROP) and clinical high-risk syndromes for psychosis (CHR-P), respectively, b) to test prediction models established in ROP and CHR-P in an independent validation sample with recent onset depression (ROD) diagnoses and c) to test for symptom expression related brain features associated with VisDys. sMRI were acquired in a training sample including 128 ROP (67 patients with VisDys), 134 CHR-P (71 patients with VisDys). Independent validation sets included 46 ROP (19 with VisDys), 124 CHR-P (68 patients with VisDys) and a sample of 256 ROD (50 patients with VisDys). Due to the heterogeneity of the CHR-P group, a clustering analysis was performed to identify proof of concept clusters of brain relevance specifically associated with symptoms in patients with and without VisDys, respectively. Both classification schemas in ROP and CHR-P presented good prediction accuracy in the independent validation samples of ROP, CHR-P and ROD, respectively. Health sciences/Diseases/Psychiatric disorders/Schizophrenia Health sciences/Diseases/Psychiatric disorders/Depression recent onset psychosis recent onset depression clusters of brain texture relevance explainable artificial intelligence Figures Figure 1 Figure 2 1. Introduction Visual dysfunctions (VisDys) are understood as subtle visual perceptual distortions of real features in the environment representing important symptoms of the wide range of perceptual impairments characterizing clinical syndromes of the psychosis spectrum [ 1 – 3 ]. Although such alterations within the visual system are often neglected in clinical practice, they are nonetheless relevant for the understanding of perception-related disease mechanisms in psychosis. The psychosis spectrum includes states of manifest psychotic syndromes such as recent onset psychosis (ROP) but also clinical high risk of psychosis (CHR-P) states, defined by ultra-high risk (UHR) or basic symptom criteria[ 4 ] besides more chronic states of psychotic disorders, e.g. recurrent episodes of schizophrenia. Of note, only about 35% of CHR-P with UHR criteria will develop into a diagnosis of manifest schizophrenia within 5 years suggesting that the CHR-P state represents a heterogeneous group in which some subgroups may share neurobiological features with ROP but other subgroups may be characterized by different neurobiological features[ 5 ]. Conversion rates for basic symptom criteria, one of which includes VisDys, may be even higher underlining the relevance of VisDys[ 4 ]. VisDys, which have first been comprehensively described by Gerd Huber[ 6 ], can manifest as perceptual abnormalities[ 7 ] regarding perceptual organization[ 8 ], contrast sensitivity[ 9 ], motion, colour, brightness, and shape including perception of human figures and emotional expressions[ 10 ]. Following a recent model, the underlying mechanisms of VisDys are understood as incorrect integration of visual information at early information processing stages within occipital networks[ 11 ] resulting in disturbed coding into neuronal signals along different visual streams, e.g. ventral and dorsal visual streams [ 12 ], respectively. Interestingly, VisDys at early stages of the disease manifest as rather hypersensitivity to visual stimuli but are suggested to turn into hyposensitivity along illness progression staying nonetheless inconsistent[ 10 ]. Previously, VisDys have been related to overt structural lesions in the occipital cortex[ 13 ]. Besides this, there is otherwise little known about structural brain tissue alterations, e.g. grey matter loss, related to VisDys. VisDys have also been linked to transdiagnostic vulnerability in mental illness presenting individual differences in aberrant structure[ 14 ] and functional connectivity of visual cortical areas of the brain[ 15 ]. We recently showed that VisDys[ 16 ] were closely related to stronger impairment of functional outcome, increased depression, and poorer quality of life, especially in CHR-P but also ROP. To expand our previous findings on functional alterations related to VisDys and taking into account the inconsistency of VisDys expression from a long-term perspective, we aim to further explore the neurobiological underpinnings related to VisDys, using structural brain MRI to predict VisDys in ROP and CHR-P from the PRONIA sample. In contrast to volumetric methods, we applied a novel approach that assesses microstructural brain texture characteristics in non-segmented brain sMRI to dive into the molecular level from another perspective which already used in another previous study of our group[ 17 ]. More specifically, texture features extracted from non-segmented brain MRI are able to reveal hidden information and explain brain complexity better than existing methods, as these features consider the inter-relations between voxels including those in different modalities (grey/white matter and cerebrospinal fluid). Based on recent own publications on brain texture capability to discriminate mental disorders[ 17 – 20 ], we hypothesized that such radiomic texture feature models capture brain alterations at a microscale level enabling the prediction of VisDys. Previous studies describing the VisDys phenomenon using follow-up study designs have pointed out their inconsistency along illness progression[ 10 ]. The distinction in the early stages of illness using brain structure data is more challenging, when symptoms and course are more heterogeneous[ 21 ]. In the present study, we therefore aimed to identify consistencies in VisDys appearance transdiagnostically across ROP and CHR-P by applying a framework that results on individual level prediction outcomes. We were further interested in the stability of these VisDys associated brain characteristics in relation to symptom load at baseline and after nine months follow-up, validating the predictability of the clinical symptom severity transdiagnostically in patients with and without VisDys, respectively. Expanding the transdiagnostic approach, we were additionally interested in the specificity of VisDys-brain texture associations in the psychosis spectrum compared to recent-onset depression (ROD) patients as an independent validation sample. 2. Method 2.1. Study Design Data were gathered as part of the EU-FP7-funded PRONIA study, a seven-center research project that aimed to optimize the use of potential biomarkers for early diagnosis and prognosis of mental diseases. 2.2. Participants The trial was registered with the German Clinical Trials Register (DRKS00005042), and local research ethics boards at each center gave their approval (Ludwig-Maximilian University Munich (ethics ID: 351 − 13), University of Basel (ethics ID: M12/99), University of Cologne (ethics ID: 13–236), University of Turku (ethics ID: 99/1810/2013), University of Bari (ethics ID: 4754), University of Milan (ethics ID: N.PROT.0001885|P|GEN/02), University of Udine (ethics ID: 67172), University of Birmingham (ethics ID: 14/WM/0019), University of Münster (ethics ID: 2016-398-b-S) and University of Düsseldorf (ethics ID: 5957A)), in accordance with the standardized recruitment and assessment protocol from the PRONIA study ( Supplementary Table S1 ). All participants provided written informed consent prior to being included in the study, as did their guardians in the case of underage participants (defined as those under the age of 18 at all locations). A comprehensive description of the study design and methodology is provided in Koutsouleris et al.[ 22 ] and in the Supplemental file . 2.3. Training and validation samples The PRONIA dataset consists of a training sample comprising patients recruited during the first project phase, and the independent validation sample comprising patients recruited during the second project phase. The training sample consisted of brain MRI data from 128 ROP (67 ROP with VisDys (ROP+), 52.3%, age and sex-adjusted to 61 ROP without VisDys (ROP-)) and 134 CHR-P (71 CHR-P with VisDys (CHR-P+), 53.0%, age and sex-adjusted to 63 CHR-P without VisDys (CHR-P-)). For independent validation of our models in age and sex-adjusted to samples to the training sample for each group, we used first, 46 ROP (19 ROP+, 41.3%, %, age and sex-adjusted to 27 ROP-) and 124 CHR-P (68 CHR-P+, 54.8%, age and sex-adjusted to 56 CHR-P-) of the PRONIA validation sample, and second, 256 ROD age and sex-adjusted with the training sample (50 ROD with VisDys (ROD+), 19.5%, age and sex-adjusted to 206 without VisDys (ROD-)) as an independent validation sample (details in Figure S1 ). 2.4. Assessments The following demographic and clinical data—age, sex, and medication exposure—were used in the current analysis: Positive and Negative Symptom Scale (PANSS)[ 23 ], the Scale of Negative Symptoms (SANS)[ 24 ], the Global Assessment of Functioning (global functioning-role scales) and the Beck Depression Inventory-II (BDI-II)[ 25 ]. To capture various facets of VisDys, 14 items were chosen from the Schizophrenia Proneness Instrument, Adult version SPI-A[ 7 , 26 ]. (see Figure S1 .2 in Supplement file, Figure S1 .3 and Table S1 .3 for more details). 2.5. Analysis 2.5.1 Preprocessing All images were visually inspected, automatically defaced, and anonymized using a Freesurfer-based script prior to data centralization. Subsequently, we used the open-source CAT12 toolbox (version r1155; http://dbm.neuro.uni-jena.de/cat12/ ), an extension of SPM12 ( http://www.fil.ion.ucl.ac.uk/spm ), to segment images into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) maps and to high-dimensionally register the segmented images to the stereotactic space of the Montreal Neurological Institute coordinates (MNI-152 space) (details in Supplemental Section 1.1 and Figure S2 ). 2.5.2 Feature extraction Using the 2D grey-level co-occurrence matrix (GLCM) computed in each cube, we were able to extract texture feature maps from non-segmented images. We extracted the texture properties of entropy, sum of entropy, difference of entropy, energy, contrast, and homogeneity based on Korda et al. [ 18 ]. We focused on the analysis in GLCM-energy and entropy, as these concluded to higher classification accuracies. Entropy measures the randomness of the texture distribution, and its inverse measure is the energy which reflects the regularity and uniformity of the texture distribution, see Figure S3 depicting a representation of energy and entropy feature maps[ 17 ]. 2.5.3 Classification framework We fed the registered energy and entropy texture feature maps into a 20x20 nested cross-validation deep learning scheme in MATLAB to train and cross-validate models to discriminate VisDys + and VisDys- in ROP and CHR-P, respectively. See Supplementary Figure S5 for further details. 2.5.4 Visualization and evaluation of heatmaps We used the LRP algorithm for multilayer neural networks, as outlined in Bach et al. [33], to determine the importance of the voxels in each class in order to localize the detected brain alterations in VisDys patients.An individual heatmap that depicts typical alterations in brain structure in general psychopathology is the algorithm's output (see Appendix A, Supplementary file for further information). Figures 1 and 2 show visualizations of the classification outcomes for both classification schemas on the holdout dataset. 2.5.5 Clustering of subjects and associations with clinical variables We calculated PANSS factor scores (positive, negative, distress, excitement, and disorganization), the SANS factor scores, psychosocial functioning (GAF), and BDI total scores, as well as VisDys scores at T0. Additionally, we calculated change scores of these variables at T1. We then calculated the Spearman rho coefficients including Bonferroni-Holm correction for multiple comparisons to investigate and model clinical symptom and outcome profiles from the mean brain relevance (heatmap) from whole brain in ROP + and ROD + groups in the independent validation at T0 and T1 hold out datasets. Due to heterogeneity of brain texture heatmaps this was not possible in CHR-P + and ROD + groups as well as CHR-P- and ROD- groups. We therefore implemented a clustering algorithm to demonstrate shared brain texture patterns across subjects to address group. Our intention was to display the heatmap of each correctly classified subject from the external independent validation sample. The independent validation sample was tested in the winner model of the training sample (see Supplementary Section 1.4 ). 3. Results 3.1. Sample characteristics. Sociodemographic and clinical characteristics of the three participant groups with and without VisDys for training (ROP, CHR-P) and validation samples (ROP, CHR-P, ROD) are presented in Table S1 .3 . Comparing training and validation samples of ROP there were no statistically significant differences between groups with and without VisDys in clinical characteristics. In CHR-P+, there were statistically significant differences between training and validation samples in GAF, and all PANSS_scores. The CHR- groups from training and validation samples were differentiated by GAF score only. In the training sample, no group differences regarding VisDys-related symptom expression were identified between ROP + vs. ROP- and CHR-P + vs. CHR-P-, respectively. Regarding the validation samples, the ROP + group showed more severe symptom expression in PANSS_negative and PANSS_excitement scores as well as GAF and SANS_anhedonia scores than the ROP- group. Additionally, higher PANSS_positive scores were observed in CHR-P + compared to CHR-P- from the validation sample. Similarly, higher PANSS_positive scores were observed in the ROD + compared to the ROD- group. In Figure S3 , it is observed that the energy feature in ROP has higher values in the limit of GM and CSF compared to the entropy in CHR-P that was higher in the outer cortical folding. In ROP, the energy feature was most informative to identify differences between ROP + and ROP-. Overall, the energy in ROP captures the inner cortex abnormalities and reflects the smoothness of the shape abnormalities, independent of the volume of the brain regions. In CHR-P, the entropy feature was most informative to identify the differences between CHR-P + and CHR-P-. In CHR-P the entropy reflects the changes of the cortical folding, independent of the volume of the brain regions (see Supplement Figure S4 for the whole preprocessing). The high entropy in the outer boundary points reflects a dense microstructural distribution which reflects to changes into cortex as well. 3.2. Classification results and localization A repeated nested pooled cross-validation classifier of a) brain energy texture maps achieved a balanced accuracy (BAC) of 84.85% for discrimination of ROP- against ROP + in the training sample. Similarly good results were achieved in the independent ROP validation sample with BAC 70.51%, Table S5.1 , and b) brain entropy texture maps achieved a BAC of 77.92% for discrimination of CHR-P + vs. CHR-P- in the training sample. In the independent CHR-P validation sample this resulted in BAC 64.08%, Table S5.1 . Regarding our third validation sample using the ROD sample, 67.73% of ROD + were classified correctly by using the energy feature ROP model for prediction of VisDys. Additionally, 62.32% of ROD + were also classified correctly using the entropy feature CHR-P model for prediction of VisDys, see Tables S5.1 and 2 . Testing the energy feature model established in ROP by using the energy feature maps established in CHR-P + vs. CHR-P- resulted in a low BAC of 53.66%, which was even lower compared to ROD energy texture feature maps (BAC 69.36%). This finding suggests common brain texture patterns between ROP and ROD. Similarly, testing the entropy feature maps established in CHR-P using the entropy feature maps of the ROP + vs. ROP- resulted in lower BACs (BAC 53.45%) compared to ROD entropy texture feature maps (BAC 66.56%). Implementation of the LRP algorithm showed that voxels with the highest contribution to the prediction of VisDys in ROP and CHR-P (i.e., those with highest relevance for the classification decision) were located in the frontal and temporal lobes (further details in Figs. 1 and 2 and Supplementary Table S5 ). The voxels that contributed significantly to the classification decision for the comparison between CHR-P + vs. CHR-P- were located in frontal and temporal lobes as well as in putamen and caudate. 3.3. Predicting clinical symptom expression at baseline and follow-up by brain texture characteristics Given the overlap between ROP and ROD regarding energy feature maps related to VisDys on the one hand, and the overlap of entropy feature maps between CHR-P and ROD related to VisDys on the other hand, the associations of the brain relevance, i.e. the voxels that contributed most to the classification decision, with clinical symptom severity at baseline and follow-up were investigated in the combined ROP/ROD validation sample and the combined CHR-P/ROD sample, respectively. We summarize the statistically significant findings here, based on symptom sum scores as described above and outcome profiles ( Supplementary file Figure S7 and S9 ). Associations of the brain relevance with symptom expression in patients presenting with VisDys : Using the energy feature maps , 44 out of 69 subjects from the combined ROP+/ROD + groups (from the validation sample) were classified correctly. The average clinical scores are represented in Tables 1 B and 1 C. Spearman correlation using the mean brain relevance from the energy feature map in combined ROP+/ROD + confirmed the associations with a) PANSS_positive, b) PANSS_disorganization, c) PANSS_excitement scores at baseline (T0) reflected by negative coefficients, i.e. lower brain relevance predicted stronger symptom expression (see Table 1 ). However, GAF score is associated positively with the brain relevance as this is captured from the energy feature. Additionally, change scores after nine-month follow-up (T1) for PANSS_disorganization were predicted with positive coefficients, i.e. higher brain relevance predicted stronger symptom expression change (improvement) over 9 months. (see Table 2 ). These associations with whole brain energy features are represented in Figure S7 A-E. Table 1 Spearman rho coefficients with p_values corrected by Bonferonni-Holm for the significant associations of whole brain mean relevance heatmaps in ROP+/ROD + at T0 with PANSS_Positive, PANSS_disorganization, PANSS_excitement and GAF (* indicates statistical significant p-values). Variable rho t P_corrected PANSS_Positive -0.58 -4.46 0.0010* PANSS_disorganization -0.56 -4.29 0.0016* PANSS_excitement -0.44 -3.11 0.0412* GAF 0.47 3.41 0.019* Table 2 Spearman rho coefficients and p_values corrected by Bonferonni-Holm for the significant association of whole brain mean relevance heatmaps in ROP+/ROD + with the change score from T0 to T1 for PANSS_excitement (* indicates statistical significant p-values). Variable rho t P_corrected PANSS_disorganization 0.42 2.39 0.015* Using the entropy feature map 62 out of 118 CHR-P+/ROD+ (from the training sample) subjects were classified correctly. Due to the heterogeneity of the CHR-P group, we applied the clustering analysis in the CHR-P/ROD validation samples resulted in 7 clusters ( Figure S10 and S12 ). On a descriptive level we found that the highest PANSS and SANS scores were identified for CHR-P + subjects in cluster 7, with the highest improvement of PANSS_negative after 9 months. The lowest BDI symptoms were grouped in cluster 5, the subjects in this group presented high remission of BDI and VisDys at T1. The lowest GAF was observed in cluster 1 at T0, with high improvement after T1. Cluster 6 presented the highest VisDys score at T0 and the highest improvement of VisDys, PANSS_distress, and anhedonia at T1. Using the mean brain relevance calculated into the clusters from entropy feature maps in combined CHR-P+/ROD + sample, we can predict the deterioration of PANSS_Positive in Cluster 5 (see Table 3 , Table S7 and Figure S9 F-H ). In cluster 5 of the CHR-P+/ROD + group lower entropy was associated with stronger deterioration of the PANSS_positive score. The subjects in cluster 5 also presented high deterioration of VisDys in the follow-up. Table 3 Spearman rho coefficients and p_values corrected by Bonferonni-Holm for the significant association of whole brain mean relevance heatmaps in CHR-P+/ROD + with the change score from T0 to T1 for PANSS_Positive in cluster 5 (* indicates statistical significant p-values). Variable rho t P_corrected PANSS_Positive (cluster 5) -0.93 -5.59 0.033* Association of the brain relevance with symptom expression in patients presenting without VisDys : In ROP-/ROD- there were no associations between whole brain relevance and symptom scores while in CHR-P-/ROD- the AP algorithm resulted in 10 clusters ( Figure S11 and S12 ). The CHR-P-/ROD- group presented higher brain heterogeneity than the CHR-P+/ROD + group. From the mean brain relevance heatmap in CHR-P-/ROD- we predicted anhedonia and asociality scores ( Table S8 ) in cluster 2 with positive coefficients, i.e. higher brain relevance predicted stronger symptom expression change. Detailed results from groups without VisDys are summarized in Tables in the Supplement. 4. Discussion Our main findings suggest that ROP + present smoother brain texture shape abnormalities in frontal and temporal lobes than ROP-, as expressed by energy texture feature maps indicating regularity and uniformity of the texture distribution. Energy in ROP reflects inner cortex features which we found to be related to global social functioning in ROP + indicated by GAF scores. Notably, we found that low energy of the brain in ROP + predicted low levels of GAF scores. This association gives us a better insight into how energy texture features are related clinically to VisDys, an association which was absent in ROP-. In contrast, CHR-P + presented sharper grey matter folding of the brain surface in frontal, temporal, and occipital areas as well as higher entropy related to putamen and caudate than CHR-P-, as expressed by entropy texture feature maps indicating the complexity of the texture distribution. Interestingly, we found that in CHR-P+/ROD + a reduction of PANSS_Positive scores from baseline to follow-up was associated with depressive symptom deterioration. Of note, entropy was higher in CHR-P+/ROD + who present higher BDI score in the follow-up compared to baseline, and this is statistically associated with the improvement of psychotic symptoms. The outer boundary points have been associated with the CHR-P identification compared to ROP in a previous study of Korda et al.[ 27 ]. This fact validates the selection of the entropy in this study, as it reflects the boundary folding, which is more dense in CHR than ROP, and especially reflects the presence of VisDys. Due to the heterogeneity of the CHR-P group, we should further investigate the role of the entropy in the outer boundary brain points with the depressive symptom severity. We further show that entropy texture was relevant for the prediction of VisDys in the CHR-P/ROD group underlining the importance of cortical folding changes in the CHR-P + group compared to ROP + group. This finding is in line with a previous report from our group in an independent sample showing that cortical boundary complexity is higher in CHR-P compared to ROP and healthy controls[ 27 ]. Notably, the resulting significant predictive models from ROP in the present study were not applicable to CHR-P and vice versa, stressing the notion that ROP and CHR-P represent distinguishable mental health conditions related to different neurobiological underpinnings. Besides these specific brain texture features, we also identified common brain patterns across ROP + and ROD + groups which were more pronounced than the common brain patterns between CHR-P + and ROD+ (see sensitivity in Table S5.1 and S5.2 ). Regarding brain texture features in probands without VisDys, some common brain abnormalities were observed across ROP-, ROD- and CHR-P-, despite a general higher heterogeneity in groups without VisDys than those with VisDys. Occipital and frontoparietal networks in the brain are known as the most relevant to the visual system [ 10 , 28 ]. Our present findings using texture features from structural brain images in contrast to intrinsic brain connectivity go beyond these previous findings[ 16 ], first, by revealing more robust and higher accuracies with equally high specificity and sensitivity for the prediction of VisDys in both ROP and CHR-P to a small-scale almost to the molecular level. These models were validated in independent samples including also a group of ROD. Note, although VisDys occur much less frequent in ROD than ROP and CHR-P, they may nonetheless indicate a certain neurobiologically based susceptibility to psychosis in individuals presenting with a depressive syndrome[ 17 ]. However, in this study the main goal was to investigate the association of the VisDys transdiagnostically with clinical symptom severity and outcome profile. Second, in line with our previous study, we found frontoparietal areas to be involved in the prediction of VisDys in all groups while occipital areas were mainly identified in CHR-P and ROD. This finding is in line with other findings from our group using radiomics texture features unrelated to VisDys, in which CHR-P presented statistically significant differences from ROP in occipital lobe[ 11 , 18 ]. Furthermore, in the present study, we found parts of the default mode and dorsal network to contribute to the prediction of VisDys in ROP. In ROP the superior longitudinal fasciculus was involved which is critically involved in visual and spatial cognition[ 29 ]. In CHR-P, the middle cerebellar peduncle and other parts of it, which convey information related to eye movement[ 30 ], corpus callosum related to unconscious vision[ 31 ] and other regions related to involuntary movements were involved. We here show that ROP presenting with VisDys were specifically impaired regarding their remission by identifying homogeneous brain texture changes, which were highly associated with PANSS_positive, disorganization and excitement scores at T0 and difference between T1 and T0, interestingly. Our present findings further support the observations from another previous study in which texture features analyses were derived from contrast images but the aspect of VisDys was neglected[ 17 ]. Similarly, in the present study we identified contrast feature clusters which contributed significantly to the classification decision, and which predicted accurately the PANSS and GAF scores in a combined ROP+/ROD + group. More specifically, in our previous paper[ 17 ], brain relevance was associated with PANSS_positive score in ROP and ROD, which is also in line with our present findings in ROP+/ROD+. These facts confirm that different texture features relate to different diagnoses, i.e. the contrast feature can predict outcomes in ROP/ROD[ 17 ], but the VisDys in ROP/ROD are identified using the energy feature. However, only ROP+/ROD + were negatively associated with PANSS scores at T0 (positive, disorganization and excitement) and positively with GAF score, which means that the lower the energy, i.e. the smoother the brain shape abnormalities at baseline, the stronger is symptom expression in patients with VisDys and disorganization scores at nine-month follow-up at T1, but lower the functionality at T0. Thus, low brain texture energy can predict worse outcome, especially related to the aspect of disorganization. We observed a similar constellation in cluster 5 of the CHR-P+/ROD + group in which subjects with low BDI score and deterioration of the BDI and VisDys score in follow-up, were associated negatively with the PANSS_positive score. This means that the sharpness of the cortical folding was associated negatively with the PANSS_score in patients with low BDI and high VisDys score. In CHR-P and ROD group the oldest subjects grouped together and presented a deterioration of PANSS scores at follow-up in both groups with or without VisDys. Also, independent of the existence of VisDys in CHR-P and ROD group the subjects with low BDI presented a deterioration of the BDI at follow-up. This is a confound of the early stage of the illness in older people and it is not related to the VisDys. In CHR-P+/ROD+, cluster 5, the highest the sharpness of the brain surface folding (entropy) the lower the PANSS_positive in patients with very low BDI which lead to worseness of VisDys in the follow-up. In addition, the change in anhedonia and asociality scores were all positively associated with brain entropy (energy) in young depressive CHR-P-/ROD-. Our present findings with respect to CHR-P are in line with our previous report from VisDys releated functional connectivity, where we found that VisDys were associated with a broader range of symptoms including impairments of quality of life, depressive symptom expression and neurocognitive impairments, i.e. visuospatial constructability, in addition to functional impairments[ 16 ]. Similarly, our present findings reflect a higher clinical heterogeneity in this group than in the ROP+/ROD + group. This constellation of findings with less clearly described symptom expression in CHR-P+/ROD + compared to ROP+/ROD + may parallel the observation that VisDys at early illness stages manifest as rather hypersensitivity reflecting a still instable state but turning into hyposensitivity along illness progression staying nonetheless inconsistent[ 10 ]. Declarations Contributions A.I.K. conceived and designed the study, feature extraction and analysis, and contributed to interpretation of the results and writing the manuscript. A.S. supported the statistical analysis and design of the study, I.M., R.L., C.A. conceived and designed the study and contributed to interpretation of the results and writing the paper. A.R. and L.H. were responsible for reprocessing of the brain MRI. U.D., G.R., L.K-I., J.K., T.L., M.R., S.R., S.S.H., R.K.R.S., C.P., C.B., F.S-L., E.M., P.B., A.B., R. U., C.D., N.K., R.L., C.A., S.B. and the PRONIA Consortium were responsible for data acquisition, support on writing the manuscript, the final approval of the version to be published and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. N.K. had full access to all the data in the study and takes responsibility for the integrity of the data. All authors provided a final approval of the version to be published. Data availability The datasets used and/or analysed during the current study will be available from the authors upon reasonable request submitted to the PRONIA Consortium. Acknowledgments Computational support and infrastructure provided by the OMICS at the University of Lübeck (Germany). Preliminary results presented at ECNP conference 2022, IEPA 2023 and SIRS 2023. Competing Interests PRONIA is a Collaboration Project funded by the European Union under the Seventh Framework Programme and grant agreement 602152. The sponsors had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript; and decision to submit the manuscript for publication. This work was funded by the German Research Foundation (DFG, grant FOR2107 DA1151/5-1, DA1151/5-2, DA1151/9-1, DA1151/10-1, DA1151/11-1 to UD; SFB/TRR 393, project grant no 521379614) and the Interdisciplinary Center for Clinical Research (IZKF) of the medical faculty of Münster (grant Dan3/022/22 to UD). N.K. received grants from the European Union (EU) during the conduct of the study and has a patent (US20160192889A1) issued. S.R. received grants from the European Commission during the conduct of the study. C.A. received non-financial support from Sunovion Pharmaceuticals, Inc. and H. Lundbeck A/S outside the submitted work. C.P. received grants from Australian National Health and the Medical Research Council during the conduct of the study and personal fees from H. Lundbeck A/S and Australia Pty Ltd outside the submitted work. R.U. received personal fees from Sunovion Pharmaceuticals, Inc. outside the submitted work. E.M. has a patent (US20160192889A1) licensed. All other authors report no biomedical financial interests or potential conflicts of interest. The PRONIA Consortium authors: The following members of the PRONIA Consortium performed the screening, recruitment, rating, examination, and follow-up of the study participants and were involved in implementing the examination protocols of the study, setting up its information technology infrastructure, and organizing the flow and quality control of the data analyzed in this study between the local study sites and the central study database: Nikolaos Koutsouleris, Anne Ruef, Lisa Hahn, Dominic B. Dwyer, Shalaila Haas, Linda A. Antonucci, Alkomiet Hasan, Claudius Hoff, Ifrah Khanyaree, Aylin Melo, Susanna Muckenhuber-Sternbauer, Yanis Köhler, Ömer Öztürk, Nora Penzel, David Popovic, Adrian Rangnick, Sebastian von Saldern, Rachele Sanfelici, Moritz Spangemacher, Ana Tupac, Maria Fernanda Urquijo-Castro, Johanna Weiske, Antonia Wosgien, and Camilla Krämer (Department ofPsychiatry and Psychotherapy, Ludwig-Maximilian-University, Germany); Lana Kambeitz-Ilankovic, Joseph Kambeitz, Julian Wenzel, Stephan Ruhrmann, Karsten Blume, Dennis Hedderich, Dominika Julkowski, Nathalie Kaiser, Thorsten Lichtenstein, Ruth Milz, Alexandra Nikolaides, Tanja Pilgram, Mauro Seves, and Martina Wassen (Department of Psychiatry and Psychotherapy, University of Cologne, Germany); Christina Andreou, Stefan Borgwardt, André Schmidt, Anita Riecher-Rössler, Laura Egloff, Fabienne Harrisberger, Ulrike Heitz, Claudia Lenz, Letizia Leanza, Amatya Mackintosh, Renata Smieskova, Erich Studerus, Anna Walter, and Sonja Widmayer (Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland); Stephen J. Wood, Rachel Upthegrove, Paris Alexandros Lalousis, Chris Day, Sian Lowri Griffiths, Mariam Iqbal, Mirabel Pelton, Pavan Mallikarjun, Alexandra Stainton, and Ashleigh Lin (Institute for Mental Health and School of Psychology, University of Birmingham, UK); Jarmo Hietala, Raimo K.R. Salokangas, Alexander Denissoff, Anu Ellilä, Tiina From, Markus Heinimaa, Tuula Ilonen, Päivi Jalo, Heikki Laurikainen, Antti Luutonen, Akseli Mäkela, Janina Paju, Henri Pesonen, Reetta-Liina Säilä, Anna Toivonen, and Otto Turtonen (Department of Psychiatry, University of Turku, Finland); Frauke Schultze-Lutter, Eva Meisenzahl, Alexandra Korda, Susanne Neufang, Christian Schmidt-Kraepelin, Henrik Rohner, Sonja Botterweck, Norman Kluthausen, Gerald Antoch, Julian Caspers, and Hans-Jörg Wittsack (Department of Psychiatry, Psychiatric University Hospital LVR/Heinrich-Heine-University Düsseldorf, University of Düsseldorf, Germany); Pierluigi Selvaggi, Alessandro Bertolino, Giuseppe Blasi, Giulio Pergola, Grazia Caforio, Leonardo Fazio, Tiziana Quarto, Barbara Gelao, Raffaella Romano, Ileana Andriola, Andrea Falsetti, Marina Barone, Roberta Passiatore, and Marina Sangiuliano (Department of Basic Medical Science, Neuroscience and Sense Organs, University of Bari Aldo Moro, Italy); Rebekka Lencer, Marian Surmann, Olga Bienek, and Udo Dannlowski (Department of Psychiatry and Psychotherapy, University of Münster, Germany); Ana Beatriz Solana, Manuela Abraham, and Timo Schirmer (GE Global Research, Inc); Paolo Brambilla, Carlo Altamura, Marika Belleri, Francesca Bottinelli, Adele Ferro, and Marta Re (Department of Neuroscience and Mental Health, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Workgroup of Paolo Brambilla, University of Milan, Italy); Emiliano Monzani and Maurizio Sberna (Programma 2000, Niguarda Hospital, Workgroup of Paolo Brambilla, University of Milan, Italy); Giampaolo Perna, Maria Nobile, and Alessandra Alciati (San Paolo Hospital, Workgroup of Paolo Brambilla, University of Milan, Italy); Armando D’Agostino and Lorenzo Del Fabro (Villa San Benedetto Menni, Albese con Cassano, Workgroup of Paolo Brambilla, University of Milan, Italy); Matteo Balestrieri, Carolina Bonivento, Giuseppe Cabras, and Franco Fabbro (Department of Medical Area, Workgroup of Paolo Brambilla, University of Udine, Italy); and Marco Garzitto and Sara Piccin (IRCCS Scientific Institute E. Medea, Polo FVG, Workgroup of Paolo Brambilla, University of Udine, Italy); Christos Pantelis (Melbourne Neuropsychiatry Centre, Department of Psychiatry, University of Melbourne & Melbourne Health, Australia), Christos Davatzikos (Department of Radiology, University of Pennsylvania School of Medicine, 3700 Hamilton Walk, Philadelphia, PA 19104). References Silverstein, S.M. and A. Lai, The phenomenology and neurobiology of visual distortions and hallucinations in schizophrenia: an update. Frontiers in Psychiatry, 2021. 12 : p. 684720. Butler, P.D., S.M. Silverstein, and S.C. Dakin, Visual Perception and Its Impairment in Schizophrenia. Biological Psychiatry, 2008. 64 (1): p. 40-47. King, D.J., et al., A review of abnormalities in the perception of visual illusions in schizophrenia. Psychonomic Bulletin & Review, 2017. 24 (3): p. 734-751. Schultze-Lutter, F., et al., EPA guidance on the early detection of clinical high risk states of psychoses. European Psychiatry, 2015. 30 (3): p. 405-416. Addington, J., et al., Progression from being at-risk to psychosis: next steps. npj Schizophrenia, 2020. 6 (1): p. 27. Schultze-Lutter, F., et al., Revisiting the Basic Symptom Concept: Toward Translating Risk Symptoms for Psychosis into Neurobiological Targets. Frontiers in Psychiatry, 2016. 7 . Silverstein, S.M., Visual Perception Disturbances in Schizophrenia: A Unified Model , in The Neuropsychopathology of Schizophrenia: Molecules, Brain Systems, Motivation, and Cognition , M. Li and W.D. Spaulding, Editors. 2016, Springer International Publishing: Cham. p. 77-132. Silverstein, S.M., et al., Perceptual organization, the disorganization syndrome, and context processing in chronic schizophrenia. Schizophrenia Research, 2000. 43 (1): p. 11-20. Cimmer, C., et al., Abnormal neurological signs, visual contrast sensitivity, and the deficit syndrome of schizophrenia. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 2006. 30 (7): p. 1225-1230. Adámek, P., V. Langová, and J. Horáček, Early-stage visual perception impairment in schizophrenia, bottom-up and back again. Schizophrenia, 2022. 8 (1): p. 27. Keane, B.P., et al., Functional dysconnectivity of visual and somatomotor networks yields a simple and robust biomarker for psychosis. Molecular Psychiatry, 2024. Hebart, M.N. and G. Hesselmann, What visual information is processed in the human dorsal stream? Journal of Neuroscience, 2012. 32 (24): p. 8107-8109. Ramenghi, L.A., et al., Visual performance and brain structures in the developing brain of pre-term infants. Early Human Development, 2010. 86 (1, Supplement): p. 73-75. Romer, A.L., et al., Replicability of structural brain alterations associated with general psychopathology: evidence from a population-representative birth cohort. Molecular psychiatry, 2021. 26 (8): p. 3839-3846. Elliott, M.L., et al., A connectome-wide functional signature of transdiagnostic risk for mental illness. Biological psychiatry, 2018. 84 (6): p. 452-459. Schwarzer, J.M., et al., The impact of visual dysfunctions in recent-onset psychosis and clinical high-risk state for psychosis. Neuropsychopharmacology, 2022. 47 (12): p. 2051-2060. Korda, A.I., et al., Brain texture as a marker of transdiagnostic clinical profiles in patients with recent-onset psychosis and depression. Nature Mental Health, 2024. 2 (1): p. 76-87. Korda Alexandra I, et al., Identification of texture MRI brain abnormalities on first-episode psychosis and clinical high-risk subjects using explainable artificial intelligence. Translational Psychiatry, 2022. 12 (1): p. 481. Korda, A.I., et al., Identification of voxel-based texture abnormalities as new biomarkers for schizophrenia and major depressive patients using layer-wise relevance propagation on deep learning decisions. Psychiatry Research: Neuroimaging, 2021. 313 : p. 111303. Hu, Y., et al., Brain Fractal Dimension and Machine Learning can predict first-episode psychosis and risk for transition to psychosis. Computers in Biology and Medicine, 2025. 193 : p. 110333. Lalousis, P.A., et al., Heterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach. Schizophrenia Bulletin, 2021. 47 (4): p. 1130-1140. Koutsouleris, N., et al., Prediction Models of Functional Outcomes for Individuals in the Clinical High-Risk State for Psychosis or With Recent-Onset Depression: A Multimodal, Multisite Machine Learning Analysis. JAMA Psychiatry, 2018. 75 (11): p. 1156-1172. Kay, S.R., A. Fiszbein, and L.A. Opler, The Positive and Negative Syndrome Scale (PANSS) for Schizophrenia. Schizophrenia Bulletin, 1987. 13 (2): p. 261-276. Andreasen, N.C., The Scale for the Assessment of Negative Symptoms (SANS): Conceptual and Theoretical Foundations. British Journal of Psychiatry, 1989. 155 (S7): p. 49-52. Beck AT, S.R., Brown GK., Beck Depression Inventory-II . 1996: San Antonio, TX: Psychological Association. Schultze-Lutter, F., et al., Schizophrenia proneness instrument, adult version (SPI-A). Rome: Giovanni Fioriti, 2007. Korda, A.I., et al., Chaos analysis of the cortical boundary for the recognition of psychosis. Journal of Psychiatry and Neuroscience, 2023. 48 (2): p. E135. Diamond, A., S.M. Silverstein, and B.P. Keane, Visual system assessment for predicting a transition to psychosis. Translational Psychiatry, 2022. 12 (1): p. 351. Podwalski, P., et al. Psychopathology and Integrity of the Superior Longitudinal Fasciculus in Deficit and Nondeficit Schizophrenia . Brain Sciences, 2022. 12 , DOI: 10.3390/brainsci12020267. Kim, S.-H. and J.-S. Kim, Eye movement abnormalities in middle cerebellar peduncle strokes. Acta Neurologica Belgica, 2019. 119 (1): p. 37-45. Alonso-Sánchez, M.F., et al., Perplexity of utterances in untreated first-episode psychosis: an ultra–high field MRI dynamic causal modelling study of the semantic network. Journal of Psychiatry and Neuroscience, 2024. 49 (4): p. E252. Additional Declarations The authors have declared there is NO conflict of interest to disclose Supplementary Files SupplementFinal.docx Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state Cite Share Download PDF Status: Published Journal Publication published 12 Feb, 2026 Read the published version in Translational Psychiatry → Version 1 posted Editorial decision: revise 08 Sep, 2025 Review # 2 received at journal 12 Aug, 2025 Review # 1 received at journal 03 Aug, 2025 Reviewer # 2 agreed at journal 28 Jul, 2025 Reviewer # 1 agreed at journal 24 Jul, 2025 Reviewers invited by journal 24 Jul, 2025 Editor assigned by journal 22 Jul, 2025 Submission checks completed at journal 22 Jul, 2025 First submitted to journal 21 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7178786","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":490447507,"identity":"03dc36d3-83cc-4a0d-8c94-d00b8e1e8954","order_by":0,"name":"Alexandra Korda","email":"data:image/png;base64,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","orcid":"","institution":"University of Lübeck","correspondingAuthor":true,"prefix":"","firstName":"Alexandra","middleName":"","lastName":"Korda","suffix":""},{"id":490447508,"identity":"d6388397-3353-4496-ba4e-ad109e9a9ea9","order_by":1,"name":"Rebekka Lencer","email":"","orcid":"https://orcid.org/0000-0003-4032-7297","institution":"University of Lübeck","correspondingAuthor":false,"prefix":"","firstName":"Rebekka","middleName":"","lastName":"Lencer","suffix":""},{"id":490447509,"identity":"970427d6-3579-41dd-a23c-edbc5d388005","order_by":2,"name":"Andreas Sprenger","email":"","orcid":"https://orcid.org/0000-0001-9255-7911","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Sprenger","suffix":""},{"id":490447510,"identity":"66e693d4-02ba-41dd-ad0a-45cb2f742181","order_by":3,"name":"Inga Meyhöfer","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Inga","middleName":"","lastName":"Meyhöfer","suffix":""},{"id":490447511,"identity":"829e29bf-6ba1-477c-8f3a-4b72edfdadd3","order_by":4,"name":"Udo Dannlowski","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Udo","middleName":"","lastName":"Dannlowski","suffix":""},{"id":490447512,"identity":"3e11ae8b-f61c-4914-be96-00158f4ebb7f","order_by":5,"name":"Georg Romer","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Georg","middleName":"","lastName":"Romer","suffix":""},{"id":490447513,"identity":"0dcb4b83-16d9-44ef-8679-7744491ef805","order_by":6,"name":"Lana Kambeitz-Ilankovic","email":"","orcid":"","institution":"University of Cologne, Faculty of Medicine and University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Lana","middleName":"","lastName":"Kambeitz-Ilankovic","suffix":""},{"id":490447514,"identity":"7ad20dcf-f9f0-420d-80d5-19e27ff09cf8","order_by":7,"name":"Joseph Kambeitz","email":"","orcid":"https://orcid.org/0000-0002-8988-3959","institution":"Faculty of Medicine and University Hospital University of Cologne, Cologne","correspondingAuthor":false,"prefix":"","firstName":"Joseph","middleName":"","lastName":"Kambeitz","suffix":""},{"id":490447515,"identity":"1df165f2-c2d2-4402-ad8b-17fc8a1964f7","order_by":8,"name":"Theresa Lichtenstein","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Theresa","middleName":"","lastName":"Lichtenstein","suffix":""},{"id":490447516,"identity":"117780f8-4621-4e42-a923-ccb1f418bf69","order_by":9,"name":"Marlene Rosen","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Marlene","middleName":"","lastName":"Rosen","suffix":""},{"id":490447517,"identity":"2d2d4ffd-3bde-44ff-809d-4328abbe1f61","order_by":10,"name":"Stephan Ruhrmann","email":"","orcid":"https://orcid.org/0000-0002-6022-2364","institution":"Faculty of Medicine and University Hospital, University of Cologne, Cologne","correspondingAuthor":false,"prefix":"","firstName":"Stephan","middleName":"","lastName":"Ruhrmann","suffix":""},{"id":490447518,"identity":"04243ed6-5efb-4fcc-9b30-1ba1d0498bad","order_by":11,"name":"Shalaila Haas","email":"","orcid":"https://orcid.org/0000-0003-1385-1050","institution":"Icahn School of Medicine at Mount Sinai","correspondingAuthor":false,"prefix":"","firstName":"Shalaila","middleName":"","lastName":"Haas","suffix":""},{"id":490447519,"identity":"51cf8731-c561-4b56-b27f-3f3a8431ef71","order_by":12,"name":"Raimo Salokangas","email":"","orcid":"","institution":"University of Turku","correspondingAuthor":false,"prefix":"","firstName":"Raimo","middleName":"","lastName":"Salokangas","suffix":""},{"id":490447520,"identity":"df73c07c-be4d-4421-bd15-d5abe4f311de","order_by":13,"name":"Christos Pantelis","email":"","orcid":"https://orcid.org/0000-0002-9565-0238","institution":"University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Christos","middleName":"","lastName":"Pantelis","suffix":""},{"id":490447521,"identity":"2e2d44c4-4c96-4a2d-8aa4-8abc24ef2f25","order_by":14,"name":"Carolina Bonivento","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Carolina","middleName":"","lastName":"Bonivento","suffix":""},{"id":490447522,"identity":"a2639567-8a1b-4056-a820-58a1b2925d9d","order_by":15,"name":"Frauke Schultze-Lutter","email":"","orcid":"https://orcid.org/0000-0003-1956-9574","institution":"Heinrich-Heine University, LVR-Klinik","correspondingAuthor":false,"prefix":"","firstName":"Frauke","middleName":"","lastName":"Schultze-Lutter","suffix":""},{"id":490447523,"identity":"c2b33df8-adfe-432b-bff9-d51f98b1c151","order_by":16,"name":"Eva Meisenzahl","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Eva","middleName":"","lastName":"Meisenzahl","suffix":""},{"id":490447524,"identity":"887f7a65-7243-4696-8987-a6a6d60009d4","order_by":17,"name":"Paolo Brambilla","email":"","orcid":"https://orcid.org/0000-0002-4021-8456","institution":"University of Milan","correspondingAuthor":false,"prefix":"","firstName":"Paolo","middleName":"","lastName":"Brambilla","suffix":""},{"id":490447525,"identity":"5ff46887-a969-4839-befa-5712ac7ef4bf","order_by":18,"name":"Alessandro Bertolino","email":"","orcid":"","institution":"Department of Translational Biomedicine and Neuroscience, University of Bari Aldo Moro, Bari, Italy.","correspondingAuthor":false,"prefix":"","firstName":"Alessandro","middleName":"","lastName":"Bertolino","suffix":""},{"id":490447526,"identity":"ef66aaed-bbbd-450c-8ec0-9caea87908cf","order_by":19,"name":"Rachel Upthegrove","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Rachel","middleName":"","lastName":"Upthegrove","suffix":""},{"id":490447527,"identity":"d83a1fa1-baf2-4b87-a414-74ea9a497d2f","order_by":20,"name":"Christos Davatzikos","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Christos","middleName":"","lastName":"Davatzikos","suffix":""},{"id":490447528,"identity":"a0821d3e-901b-4ddd-9b97-ca8136363b08","order_by":21,"name":"Nikolaos Koutsouleris","email":"","orcid":"https://orcid.org/0000-0001-6825-6262","institution":"Department of Psychiatry and Psychotherapy, Ludwig-Maximilians-University, Munich, Germany.","correspondingAuthor":false,"prefix":"","firstName":"Nikolaos","middleName":"","lastName":"Koutsouleris","suffix":""},{"id":490447529,"identity":"27608568-123a-4154-834e-b3082aedcef6","order_by":22,"name":"Stefan Borgwardt","email":"","orcid":"https://orcid.org/0000-0002-5792-3987","institution":"University of Lübeck","correspondingAuthor":false,"prefix":"","firstName":"Stefan","middleName":"","lastName":"Borgwardt","suffix":""},{"id":490447530,"identity":"f718597e-0410-4612-9327-bd18bade0c74","order_by":23,"name":"Christina Andreou","email":"","orcid":"https://orcid.org/0000-0002-6656-9043","institution":"UKSH Luebeck","correspondingAuthor":false,"prefix":"","firstName":"Christina","middleName":"","lastName":"Andreou","suffix":""}],"badges":[],"createdAt":"2025-07-21 14:55:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7178786/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7178786/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41398-026-03840-x","type":"published","date":"2026-02-12T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":87804514,"identity":"8aa2949b-f914-4849-90fb-78a147f3e537","added_by":"auto","created_at":"2025-07-29 08:17:43","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":372019,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentation of the voxels that contribute most to the classification decision ROP+/ROD+ based on energy texture features. Different colors represent the fraction of the voxels for each region that are defined as significant.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7178786/v1/63e6646f78a7a1cbf8a6ce99.png"},{"id":87804516,"identity":"0ee82fff-789d-44c8-9fae-88fd8e4c2e7d","added_by":"auto","created_at":"2025-07-29 08:17:43","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":320291,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentation of the voxels that contribute most to the classification decision CHR-P+/ROD+ based on entropy texture features. Different colors represent the fraction of the voxels for each region that are defined as significant.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7178786/v1/edfe9aaf0f75f653ac26ead4.png"},{"id":103119525,"identity":"fc71235e-4708-4cf2-8747-968a80ff6c1a","added_by":"auto","created_at":"2026-02-21 08:10:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1842906,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7178786/v1/1c185ee2-1375-4450-9e89-ce25313c092b.pdf"},{"id":87804517,"identity":"9d429ffd-d2e9-4b9e-ab27-cd39586440e2","added_by":"auto","created_at":"2025-07-29 08:17:43","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2822541,"visible":true,"origin":"","legend":"Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state","description":"","filename":"SupplementFinal.docx","url":"https://assets-eu.researchsquare.com/files/rs-7178786/v1/83f037dc74f9f6a12191a4e9.docx"}],"financialInterests":"The authors have declared there is \u003cb\u003eNO\u003c/b\u003e conflict of interest to disclose","formattedTitle":"Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eVisual dysfunctions (VisDys) are understood as subtle visual perceptual distortions of real features in the environment representing important symptoms of the wide range of perceptual impairments characterizing clinical syndromes of the psychosis spectrum [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Although such alterations within the visual system are often neglected in clinical practice, they are nonetheless relevant for the understanding of perception-related disease mechanisms in psychosis. The psychosis spectrum includes states of manifest psychotic syndromes such as recent onset psychosis (ROP) but also clinical high risk of psychosis (CHR-P) states, defined by ultra-high risk (UHR) or basic symptom criteria[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] besides more chronic states of psychotic disorders, e.g. recurrent episodes of schizophrenia. Of note, only about 35% of CHR-P with UHR criteria will develop into a diagnosis of manifest schizophrenia within 5 years suggesting that the CHR-P state represents a heterogeneous group in which some subgroups may share neurobiological features with ROP but other subgroups may be characterized by different neurobiological features[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Conversion rates for basic symptom criteria, one of which includes VisDys, may be even higher underlining the relevance of VisDys[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. VisDys, which have first been comprehensively described by Gerd Huber[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], can manifest as perceptual abnormalities[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e] regarding perceptual organization[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], contrast sensitivity[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], motion, colour, brightness, and shape including perception of human figures and emotional expressions[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eFollowing a recent model, the underlying mechanisms of VisDys are understood as incorrect integration of visual information at early information processing stages within occipital networks[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] resulting in disturbed coding into neuronal signals along different visual streams, e.g. ventral and dorsal visual streams [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], respectively. Interestingly, VisDys at early stages of the disease manifest as rather hypersensitivity to visual stimuli but are suggested to turn into hyposensitivity along illness progression staying nonetheless inconsistent[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Previously, VisDys have been related to overt structural lesions in the occipital cortex[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Besides this, there is otherwise little known about structural brain tissue alterations, e.g. grey matter loss, related to VisDys. VisDys have also been linked to transdiagnostic vulnerability in mental illness presenting individual differences in aberrant structure[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] and functional connectivity of visual cortical areas of the brain[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. We recently showed that VisDys[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] were closely related to stronger impairment of functional outcome, increased depression, and poorer quality of life, especially in CHR-P but also ROP.\u003c/p\u003e\u003cp\u003eTo expand our previous findings on functional alterations related to VisDys and taking into account the inconsistency of VisDys expression from a long-term perspective, we aim to further explore the neurobiological underpinnings related to VisDys, using structural brain MRI to predict VisDys in ROP and CHR-P from the PRONIA sample. In contrast to volumetric methods, we applied a novel approach that assesses microstructural brain texture characteristics in non-segmented brain sMRI to dive into the molecular level from another perspective which already used in another previous study of our group[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. More specifically, texture features extracted from non-segmented brain MRI are able to reveal hidden information and explain brain complexity better than existing methods, as these features consider the inter-relations between voxels including those in different modalities (grey/white matter and cerebrospinal fluid). Based on recent own publications on brain texture capability to discriminate mental disorders[\u003cspan additionalcitationids=\"CR18 CR19\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], we hypothesized that such radiomic texture feature models capture brain alterations at a microscale level enabling the prediction of VisDys. Previous studies describing the VisDys phenomenon using follow-up study designs have pointed out their inconsistency along illness progression[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. The distinction in the early stages of illness using brain structure data is more challenging, when symptoms and course are more heterogeneous[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. In the present study, we therefore aimed to identify consistencies in VisDys appearance transdiagnostically across ROP and CHR-P by applying a framework that results on individual level prediction outcomes. We were further interested in the stability of these VisDys associated brain characteristics in relation to symptom load at baseline and after nine months follow-up, validating the predictability of the clinical symptom severity transdiagnostically in patients with and without VisDys, respectively. Expanding the transdiagnostic approach, we were additionally interested in the specificity of VisDys-brain texture associations in the psychosis spectrum compared to recent-onset depression (ROD) patients as an independent validation sample.\u003c/p\u003e"},{"header":"2. Method","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Study Design\u003c/h2\u003e\u003cp\u003eData were gathered as part of the EU-FP7-funded PRONIA study, a seven-center research project that aimed to optimize the use of potential biomarkers for early diagnosis and prognosis of mental diseases.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Participants\u003c/h2\u003e\u003cp\u003eThe trial was registered with the German Clinical Trials Register (DRKS00005042), and local research ethics boards at each center gave their approval (Ludwig-Maximilian University Munich (ethics ID: 351\u0026thinsp;\u0026minus;\u0026thinsp;13), University of Basel (ethics ID: M12/99), University of Cologne (ethics ID: 13\u0026ndash;236), University of Turku (ethics ID: 99/1810/2013), University of Bari (ethics ID: 4754), University of Milan (ethics ID: N.PROT.0001885|P|GEN/02), University of Udine (ethics ID: 67172), University of Birmingham (ethics ID: 14/WM/0019), University of M\u0026uuml;nster (ethics ID: 2016-398-b-S) and University of D\u0026uuml;sseldorf (ethics ID: 5957A)), in accordance with the standardized recruitment and assessment protocol from the PRONIA study (\u003cb\u003eSupplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e). All participants provided written informed consent prior to being included in the study, as did their guardians in the case of underage participants (defined as those under the age of 18 at all locations). A comprehensive description of the study design and methodology is provided in Koutsouleris et al.[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] and in the \u003cb\u003eSupplemental file\u003c/b\u003e.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Training and validation samples\u003c/h2\u003e\u003cp\u003eThe PRONIA dataset consists of a training sample comprising patients recruited during the first project phase, and the independent validation sample comprising patients recruited during the second project phase. The training sample consisted of brain MRI data from 128 ROP (67 ROP with VisDys (ROP+), 52.3%, age and sex-adjusted to 61 ROP without VisDys (ROP-)) and 134 CHR-P (71 CHR-P with VisDys (CHR-P+), 53.0%, age and sex-adjusted to 63 CHR-P without VisDys (CHR-P-)). For independent validation of our models in age and sex-adjusted to samples to the training sample for each group, we used first, 46 ROP (19 ROP+, 41.3%, %, age and sex-adjusted to 27 ROP-) and 124 CHR-P (68 CHR-P+, 54.8%, age and sex-adjusted to 56 CHR-P-) of the PRONIA validation sample, and second, 256 ROD age and sex-adjusted with the training sample (50 ROD with VisDys (ROD+), 19.5%, age and sex-adjusted to 206 without VisDys (ROD-)) as an independent validation sample (details in \u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Assessments\u003c/h2\u003e\u003cp\u003eThe following demographic and clinical data\u0026mdash;age, sex, and medication exposure\u0026mdash;were used in the current analysis: Positive and Negative Symptom Scale (PANSS)[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], the Scale of Negative Symptoms (SANS)[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], the Global Assessment of Functioning (global functioning-role scales) and the Beck Depression Inventory-II (BDI-II)[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. To capture various facets of VisDys, 14 items were chosen from the Schizophrenia Proneness Instrument, Adult version SPI-A[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. (see \u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.2\u003c/b\u003e in \u003cb\u003eSupplement file, Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.3 and Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.3\u003c/b\u003e for more details).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003e2.5. Analysis\u003c/h2\u003e\u003cdiv id=\"Sec8\" class=\"Section3\"\u003e\u003ch2\u003e2.5.1 Preprocessing\u003c/h2\u003e\u003cp\u003eAll images were visually inspected, automatically defaced, and anonymized using a Freesurfer-based script prior to data centralization. Subsequently, we used the open-source CAT12 toolbox (version r1155; \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dbm.neuro.uni-jena.de/cat12/\u003c/span\u003e\u003cspan address=\"http://dbm.neuro.uni-jena.de/cat12/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), an extension of SPM12 (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.fil.ion.ucl.ac.uk/spm\u003c/span\u003e\u003cspan address=\"http://www.fil.ion.ucl.ac.uk/spm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), to segment images into gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) maps and to high-dimensionally register the segmented images to the stereotactic space of the Montreal Neurological Institute coordinates (MNI-152 space) (details in \u003cb\u003eSupplemental Section 1.1\u003c/b\u003e and \u003cb\u003eFigure S2\u003c/b\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section3\"\u003e\u003ch2\u003e2.5.2 Feature extraction\u003c/h2\u003e\u003cp\u003eUsing the 2D grey-level co-occurrence matrix (GLCM) computed in each cube, we were able to extract texture feature maps from non-segmented images. We extracted the texture properties of entropy, sum of entropy, difference of entropy, energy, contrast, and homogeneity based on Korda et al. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. We focused on the analysis in GLCM-energy and entropy, as these concluded to higher classification accuracies. Entropy measures the randomness of the texture distribution, and its inverse measure is the energy which reflects the regularity and uniformity of the texture distribution, see \u003cb\u003eFigure S3\u003c/b\u003e depicting a representation of energy and entropy feature maps[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec10\" class=\"Section3\"\u003e\u003ch2\u003e2.5.3 Classification framework\u003c/h2\u003e\u003cp\u003eWe fed the registered energy and entropy texture feature maps into a 20x20 nested cross-validation deep learning scheme in MATLAB to train and cross-validate models to discriminate VisDys\u0026thinsp;+\u0026thinsp;and VisDys- in ROP and CHR-P, respectively. See \u003cb\u003eSupplementary Figure S5\u003c/b\u003e for further details.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec11\" class=\"Section3\"\u003e\u003ch2\u003e2.5.4 Visualization and evaluation of heatmaps\u003c/h2\u003e\u003cp\u003eWe used the LRP algorithm for multilayer neural networks, as outlined in Bach et al. [33], to determine the importance of the voxels in each class in order to localize the detected brain alterations in VisDys patients.An individual heatmap that depicts typical alterations in brain structure in general psychopathology is the algorithm's output (see Appendix A, Supplementary file for further information). Figures\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e show visualizations of the classification outcomes for both classification schemas on the holdout dataset.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section3\"\u003e\u003ch2\u003e2.5.5 Clustering of subjects and associations with clinical variables\u003c/h2\u003e\u003cp\u003eWe calculated PANSS factor scores (positive, negative, distress, excitement, and disorganization), the SANS factor scores, psychosocial functioning (GAF), and BDI total scores, as well as VisDys scores at T0. Additionally, we calculated change scores of these variables at T1. We then calculated the Spearman rho coefficients including Bonferroni-Holm correction for multiple comparisons to investigate and model clinical symptom and outcome profiles from the mean brain relevance (heatmap) from whole brain in ROP\u0026thinsp;+\u0026thinsp;and ROD\u0026thinsp;+\u0026thinsp;groups in the independent validation at T0 and T1 hold out datasets. Due to heterogeneity of brain texture heatmaps this was not possible in CHR-P\u0026thinsp;+\u0026thinsp;and ROD\u0026thinsp;+\u0026thinsp;groups as well as CHR-P- and ROD- groups. We therefore implemented a clustering algorithm to demonstrate shared brain texture patterns across subjects to address group. Our intention was to display the heatmap of each correctly classified subject from the external independent validation sample. The independent validation sample was tested in the winner model of the training sample (see \u003cb\u003eSupplementary Section 1.4\u003c/b\u003e).\u003c/p\u003e\u003c/div\u003e\u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Sample characteristics.\u003c/h2\u003e\u003cp\u003eSociodemographic and clinical characteristics of the three participant groups with and without VisDys for training (ROP, CHR-P) and validation samples (ROP, CHR-P, ROD) are presented in \u003cb\u003eTable \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.3\u003c/b\u003e. Comparing training and validation samples of ROP there were no statistically significant differences between groups with and without VisDys in clinical characteristics. In CHR-P+, there were statistically significant differences between training and validation samples in GAF, and all PANSS_scores. The CHR- groups from training and validation samples were differentiated by GAF score only. In the training sample, no group differences regarding VisDys-related symptom expression were identified between ROP\u0026thinsp;+\u0026thinsp;vs. ROP- and CHR-P\u0026thinsp;+\u0026thinsp;vs. CHR-P-, respectively. Regarding the validation samples, the ROP\u0026thinsp;+\u0026thinsp;group showed more severe symptom expression in PANSS_negative and PANSS_excitement scores as well as GAF and SANS_anhedonia scores than the ROP- group. Additionally, higher PANSS_positive scores were observed in CHR-P\u0026thinsp;+\u0026thinsp;compared to CHR-P- from the validation sample. Similarly, higher PANSS_positive scores were observed in the ROD\u0026thinsp;+\u0026thinsp;compared to the ROD- group.\u003c/p\u003e\u003cp\u003eIn \u003cb\u003eFigure S3\u003c/b\u003e, it is observed that the energy feature in ROP has higher values in the limit of GM and CSF compared to the entropy in CHR-P that was higher in the outer cortical folding. In ROP, the energy feature was most informative to identify differences between ROP\u0026thinsp;+\u0026thinsp;and ROP-. Overall, the energy in ROP captures the inner cortex abnormalities and reflects the smoothness of the shape abnormalities, independent of the volume of the brain regions. In CHR-P, the entropy feature was most informative to identify the differences between CHR-P\u0026thinsp;+\u0026thinsp;and CHR-P-. In CHR-P the entropy reflects the changes of the cortical folding, independent of the volume of the brain regions (see \u003cb\u003eSupplement Figure S4\u003c/b\u003e for the whole preprocessing). The high entropy in the outer boundary points reflects a dense microstructural distribution which reflects to changes into cortex as well.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\u003ch2\u003e3.2. Classification results and localization\u003c/h2\u003e\u003cp\u003eA repeated nested pooled cross-validation classifier of a) \u003cb\u003ebrain energy texture maps\u003c/b\u003e achieved a balanced accuracy (BAC) of 84.85% for discrimination of ROP- against ROP\u0026thinsp;+\u0026thinsp;in the training sample. Similarly good results were achieved in the independent ROP validation sample with BAC 70.51%, \u003cb\u003eTable S5.1\u003c/b\u003e, and b) \u003cb\u003ebrain entropy texture maps\u003c/b\u003e achieved a BAC of 77.92% for discrimination of CHR-P\u0026thinsp;+\u0026thinsp;vs. CHR-P- in the training sample. In the independent CHR-P validation sample this resulted in BAC 64.08%, \u003cb\u003eTable S5.1\u003c/b\u003e. Regarding our third validation sample using the ROD sample, 67.73% of ROD\u0026thinsp;+\u0026thinsp;were classified correctly by using the energy feature ROP model for prediction of VisDys. Additionally, 62.32% of ROD\u0026thinsp;+\u0026thinsp;were also classified correctly using the entropy feature CHR-P model for prediction of VisDys, see \u003cb\u003eTables S5.1\u003c/b\u003e and \u003cb\u003e2\u003c/b\u003e. Testing the energy feature model established in ROP by using the energy feature maps established in CHR-P\u0026thinsp;+\u0026thinsp;vs. CHR-P- resulted in a low BAC of 53.66%, which was even lower compared to ROD energy texture feature maps (BAC 69.36%). This finding suggests common brain texture patterns between ROP and ROD. Similarly, testing the entropy feature maps established in CHR-P using the entropy feature maps of the ROP\u0026thinsp;+\u0026thinsp;vs. ROP- resulted in lower BACs (BAC 53.45%) compared to ROD entropy texture feature maps (BAC 66.56%).\u003c/p\u003e\u003cp\u003eImplementation of the LRP algorithm showed that voxels with the highest contribution to the prediction of VisDys in ROP and CHR-P (i.e., those with highest relevance for the classification decision) were located in the frontal and temporal lobes (further details in Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and \u003cb\u003eSupplementary Table S5\u003c/b\u003e). The voxels that contributed significantly to the classification decision for the comparison between CHR-P\u0026thinsp;+\u0026thinsp;vs. CHR-P- were located in frontal and temporal lobes as well as in putamen and caudate.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\u003ch2\u003e3.3. Predicting clinical symptom expression at baseline and follow-up by brain texture characteristics\u003c/h2\u003e\u003cp\u003eGiven the overlap between ROP and ROD regarding energy feature maps related to VisDys on the one hand, and the overlap of entropy feature maps between CHR-P and ROD related to VisDys on the other hand, the associations of the brain relevance, i.e. the voxels that contributed most to the classification decision, with clinical symptom severity at baseline and follow-up were investigated in the combined ROP/ROD validation sample and the combined CHR-P/ROD sample, respectively. We summarize the statistically significant findings here, based on symptom sum scores as described above and outcome profiles (\u003cb\u003eSupplementary file Figure S7 and S9\u003c/b\u003e).\u003c/p\u003e\u003cp\u003e\u003cem\u003eAssociations of the brain relevance with symptom expression in patients presenting with VisDys\u003c/em\u003e:\u003c/p\u003e\u003cp\u003eUsing the \u003cb\u003eenergy feature maps\u003c/b\u003e, 44 out of 69 subjects from the combined ROP+/ROD\u0026thinsp;+\u0026thinsp;groups (from the validation sample) were classified correctly. The average clinical scores are represented in Tables\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC. Spearman correlation using the mean brain relevance from the \u003cb\u003eenergy feature map\u003c/b\u003e in combined ROP+/ROD\u0026thinsp;+\u0026thinsp;confirmed the associations with a) PANSS_positive, b) PANSS_disorganization, c) PANSS_excitement scores at baseline (T0) reflected by negative coefficients, i.e. lower brain relevance predicted stronger symptom expression (see Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). However, GAF score is associated positively with the brain relevance as this is captured from the energy feature. Additionally, change scores after nine-month follow-up (T1) for PANSS_disorganization were predicted with positive coefficients, i.e. higher brain relevance predicted stronger symptom expression change (improvement) over 9 months. (see Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These associations with whole brain energy features are represented in \u003cb\u003eFigure S7 A-E.\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSpearman rho coefficients with p_values corrected by Bonferonni-Holm for the significant associations of whole brain mean relevance heatmaps in ROP+/ROD\u0026thinsp;+\u0026thinsp;at T0 with PANSS_Positive, PANSS_disorganization, PANSS_excitement and GAF (* indicates statistical significant p-values).\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003erho\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003et\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP_corrected\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePANSS_Positive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.58\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e-4.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.0010*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePANSS_disorganization\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e-4.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.0016*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePANSS_excitement\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-0.44\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e-3.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.0412*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGAF\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.47\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.019*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSpearman rho coefficients and p_values corrected by Bonferonni-Holm for the significant association of whole brain mean relevance heatmaps in ROP+/ROD\u0026thinsp;+\u0026thinsp;with the change score from T0 to T1 for PANSS_excitement (* indicates statistical significant p-values).\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003erho\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003et\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP_corrected\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePANSS_disorganization\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e2.39\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.015*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eUsing the \u003cb\u003eentropy feature map\u003c/b\u003e 62 out of 118 CHR-P+/ROD+ (from the training sample) subjects were classified correctly. Due to the heterogeneity of the CHR-P group, we applied the clustering analysis in the CHR-P/ROD validation samples resulted in 7 clusters (\u003cb\u003eFigure S10\u003c/b\u003e and \u003cb\u003eS12\u003c/b\u003e). On a descriptive level we found that the highest PANSS and SANS scores were identified for CHR-P\u0026thinsp;+\u0026thinsp;subjects in cluster 7, with the highest improvement of PANSS_negative after 9 months. The lowest BDI symptoms were grouped in cluster 5, the subjects in this group presented high remission of BDI and VisDys at T1. The lowest GAF was observed in cluster 1 at T0, with high improvement after T1. Cluster 6 presented the highest VisDys score at T0 and the highest improvement of VisDys, PANSS_distress, and anhedonia at T1. Using the mean brain relevance calculated into the clusters from entropy feature maps in combined CHR-P+/ROD\u0026thinsp;+\u0026thinsp;sample, we can predict the deterioration of PANSS_Positive in Cluster 5 (see Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, \u003cb\u003eTable S7 and Figure S9 F-H\u003c/b\u003e). In cluster 5 of the CHR-P+/ROD\u0026thinsp;+\u0026thinsp;group lower entropy was associated with stronger deterioration of the PANSS_positive score. The subjects in cluster 5 also presented high deterioration of VisDys in the follow-up.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eSpearman rho coefficients and p_values corrected by Bonferonni-Holm for the significant association of whole brain mean relevance heatmaps in CHR-P+/ROD\u0026thinsp;+\u0026thinsp;with the change score from T0 to T1 for PANSS_Positive in cluster 5 (* indicates statistical significant p-values).\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariable\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003erho\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003et\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eP_corrected\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePANSS_Positive (cluster 5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-5.59\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.033*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003cem\u003eAssociation of the brain relevance with symptom expression in patients presenting without VisDys\u003c/em\u003e:\u003c/p\u003e\u003cp\u003eIn ROP-/ROD- there were no associations between whole brain relevance and symptom scores while in CHR-P-/ROD- the AP algorithm resulted in 10 clusters (\u003cb\u003eFigure S11\u003c/b\u003e and \u003cb\u003eS12\u003c/b\u003e). The CHR-P-/ROD- group presented higher brain heterogeneity than the CHR-P+/ROD\u0026thinsp;+\u0026thinsp;group. From the mean brain relevance heatmap in CHR-P-/ROD- we predicted anhedonia and asociality scores (\u003cb\u003eTable S8\u003c/b\u003e) in cluster 2 with positive coefficients, i.e. higher brain relevance predicted stronger symptom expression change. Detailed results from groups without VisDys are summarized in Tables in the Supplement.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eOur main findings suggest that ROP\u0026thinsp;+\u0026thinsp;present smoother brain texture shape abnormalities in frontal and temporal lobes than ROP-, as expressed by energy texture feature maps indicating regularity and uniformity of the texture distribution. Energy in ROP reflects inner cortex features which we found to be related to global social functioning in ROP\u0026thinsp;+\u0026thinsp;indicated by GAF scores. Notably, we found that low energy of the brain in ROP\u0026thinsp;+\u0026thinsp;predicted low levels of GAF scores. This association gives us a better insight into how energy texture features are related clinically to VisDys, an association which was absent in ROP-. In contrast, CHR-P\u0026thinsp;+\u0026thinsp;presented sharper grey matter folding of the brain surface in frontal, temporal, and occipital areas as well as higher entropy related to putamen and caudate than CHR-P-, as expressed by entropy texture feature maps indicating the complexity of the texture distribution. Interestingly, we found that in CHR-P+/ROD\u0026thinsp;+\u0026thinsp;a reduction of PANSS_Positive scores from baseline to follow-up was associated with depressive symptom deterioration. Of note, entropy was higher in CHR-P+/ROD\u0026thinsp;+\u0026thinsp;who present higher BDI score in the follow-up compared to baseline, and this is statistically associated with the improvement of psychotic symptoms. The outer boundary points have been associated with the CHR-P identification compared to ROP in a previous study of Korda et al.[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. This fact validates the selection of the entropy in this study, as it reflects the boundary folding, which is more dense in CHR than ROP, and especially reflects the presence of VisDys. Due to the heterogeneity of the CHR-P group, we should further investigate the role of the entropy in the outer boundary brain points with the depressive symptom severity. We further show that entropy texture was relevant for the prediction of VisDys in the CHR-P/ROD group underlining the importance of cortical folding changes in the CHR-P\u0026thinsp;+\u0026thinsp;group compared to ROP\u0026thinsp;+\u0026thinsp;group. This finding is in line with a previous report from our group in an independent sample showing that cortical boundary complexity is higher in CHR-P compared to ROP and healthy controls[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. Notably, the resulting significant predictive models from ROP in the present study were not applicable to CHR-P and vice versa, stressing the notion that ROP and CHR-P represent distinguishable mental health conditions related to different neurobiological underpinnings. Besides these specific brain texture features, we also identified common brain patterns across ROP\u0026thinsp;+\u0026thinsp;and ROD\u0026thinsp;+\u0026thinsp;groups which were more pronounced than the common brain patterns between CHR-P\u0026thinsp;+\u0026thinsp;and ROD+ (see sensitivity in \u003cb\u003eTable S5.1 and S5.2\u003c/b\u003e). Regarding brain texture features in probands without VisDys, some common brain abnormalities were observed across ROP-, ROD- and CHR-P-, despite a general higher heterogeneity in groups without VisDys than those with VisDys.\u003c/p\u003e\u003cp\u003eOccipital and frontoparietal networks in the brain are known as the most relevant to the visual system [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eOur present findings using texture features from structural brain images in contrast to intrinsic brain connectivity go beyond these previous findings[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], first, by revealing more robust and higher accuracies with equally high specificity and sensitivity for the prediction of VisDys in both ROP and CHR-P to a small-scale almost to the molecular level. These models were validated in independent samples including also a group of ROD. Note, although VisDys occur much less frequent in ROD than ROP and CHR-P, they may nonetheless indicate a certain neurobiologically based susceptibility to psychosis in individuals presenting with a depressive syndrome[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. However, in this study the main goal was to investigate the association of the VisDys transdiagnostically with clinical symptom severity and outcome profile. Second, in line with our previous study, we found frontoparietal areas to be involved in the prediction of VisDys in all groups while occipital areas were mainly identified in CHR-P and ROD. This finding is in line with other findings from our group using radiomics texture features unrelated to VisDys, in which CHR-P presented statistically significant differences from ROP in occipital lobe[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Furthermore, in the present study, we found parts of the default mode and dorsal network to contribute to the prediction of VisDys in ROP. In ROP the superior longitudinal fasciculus was involved which is critically involved in visual and spatial cognition[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. In CHR-P, the middle cerebellar peduncle and other parts of it, which convey information related to eye movement[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], corpus callosum related to unconscious vision[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] and other regions related to involuntary movements were involved. We here show that ROP presenting with VisDys were specifically impaired regarding their remission by identifying homogeneous brain texture changes, which were highly associated with PANSS_positive, disorganization and excitement scores at T0 and difference between T1 and T0, interestingly. Our present findings further support the observations from another previous study in which texture features analyses were derived from contrast images but the aspect of VisDys was neglected[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Similarly, in the present study we identified contrast feature clusters which contributed significantly to the classification decision, and which predicted accurately the PANSS and GAF scores in a combined ROP+/ROD\u0026thinsp;+\u0026thinsp;group.\u003c/p\u003e\u003cp\u003eMore specifically, in our previous paper[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], brain relevance was associated with PANSS_positive score in ROP and ROD, which is also in line with our present findings in ROP+/ROD+. These facts confirm that different texture features relate to different diagnoses, i.e. the contrast feature can predict outcomes in ROP/ROD[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], but the VisDys in ROP/ROD are identified using the energy feature. However, only ROP+/ROD\u0026thinsp;+\u0026thinsp;were negatively associated with PANSS scores at T0 (positive, disorganization and excitement) and positively with GAF score, which means that the lower the energy, i.e. the smoother the brain shape abnormalities at baseline, the stronger is symptom expression in patients with VisDys and disorganization scores at nine-month follow-up at T1, but lower the functionality at T0. Thus, low brain texture energy can predict worse outcome, especially related to the aspect of disorganization. We observed a similar constellation in cluster 5 of the CHR-P+/ROD\u0026thinsp;+\u0026thinsp;group in which subjects with low BDI score and deterioration of the BDI and VisDys score in follow-up, were associated negatively with the PANSS_positive score. This means that the sharpness of the cortical folding was associated negatively with the PANSS_score in patients with low BDI and high VisDys score. In CHR-P and ROD group the oldest subjects grouped together and presented a deterioration of PANSS scores at follow-up in both groups with or without VisDys. Also, independent of the existence of VisDys in CHR-P and ROD group the subjects with low BDI presented a deterioration of the BDI at follow-up. This is a confound of the early stage of the illness in older people and it is not related to the VisDys. In CHR-P+/ROD+, cluster 5, the highest the sharpness of the brain surface folding (entropy) the lower the PANSS_positive in patients with very low BDI which lead to worseness of VisDys in the follow-up. In addition, the change in anhedonia and asociality scores were all positively associated with brain entropy (energy) in young depressive CHR-P-/ROD-.\u003c/p\u003e\u003cp\u003eOur present findings with respect to CHR-P are in line with our previous report from VisDys releated functional connectivity, where we found that VisDys were associated with a broader range of symptoms including impairments of quality of life, depressive symptom expression and neurocognitive impairments, i.e. visuospatial constructability, in addition to functional impairments[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Similarly, our present findings reflect a higher clinical heterogeneity in this group than in the ROP+/ROD\u0026thinsp;+\u0026thinsp;group. This constellation of findings with less clearly described symptom expression in CHR-P+/ROD\u0026thinsp;+\u0026thinsp;compared to ROP+/ROD\u0026thinsp;+\u0026thinsp;may parallel the observation that VisDys at early illness stages manifest as rather hypersensitivity reflecting a still instable state but turning into hyposensitivity along illness progression staying nonetheless inconsistent[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eContributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA.I.K. conceived and designed the study, feature extraction and analysis, and contributed to interpretation of the results and writing the manuscript. A.S. supported the statistical analysis and design of the study, I.M., R.L., C.A. conceived and designed the study and contributed to interpretation of the results and writing the paper. A.R. and L.H. were responsible for reprocessing of the brain MRI. U.D., G.R., L.K-I., J.K., T.L., M.R., S.R., S.S.H., R.K.R.S., C.P., C.B., F.S-L., E.M., P.B., A.B., R. U., C.D., N.K., R.L., C.A., \u0026nbsp;S.B. and the PRONIA Consortium were responsible for data acquisition, support on writing the manuscript, the final approval of the version to be published and agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. N.K. had full access to all the data in the study and takes responsibility for the integrity of the data. All authors provided a final approval of the version to be published.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study will be available from the authors upon reasonable request submitted to the PRONIA Consortium.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eComputational support and infrastructure provided by the OMICS at the University of Lübeck (Germany). Preliminary results presented at ECNP conference 2022, IEPA 2023 and SIRS 2023.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePRONIA is a Collaboration Project funded by the European Union under the Seventh Framework Programme and grant agreement 602152. The sponsors had no role in the design and conduct of the study; collection, management, analysis and interpretation of the data; preparation, review or approval of the manuscript; and decision to submit the manuscript for publication.\u003c/p\u003e\n\u003cp\u003eThis work was funded by the German Research Foundation (DFG, grant FOR2107 DA1151/5-1, DA1151/5-2, DA1151/9-1, DA1151/10-1, DA1151/11-1 to UD; SFB/TRR 393, project grant no 521379614) and the Interdisciplinary Center for Clinical Research (IZKF) of the medical faculty of Münster (grant Dan3/022/22 to UD). N.K. received grants from the European Union (EU) during the conduct of the study and has a patent (US20160192889A1) issued. S.R. received grants from the European Commission during the conduct of the study. C.A. received non-financial support from Sunovion Pharmaceuticals, Inc. and H. Lundbeck A/S outside the submitted work. C.P. received grants from Australian National Health and the Medical Research Council during the conduct of the study and personal fees from H. Lundbeck A/S and Australia Pty Ltd outside the submitted work. R.U. received personal fees from Sunovion Pharmaceuticals, Inc. outside the submitted work. E.M. has a patent (US20160192889A1) licensed. All other authors report no biomedical financial interests or potential conflicts of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe PRONIA Consortium authors:\u003c/strong\u003e The following members of the PRONIA Consortium performed the screening, recruitment, rating, examination, and follow-up of the study participants and were involved in implementing the examination protocols of the study, setting up its information technology infrastructure, and organizing the flow and quality control of the data analyzed in this study between the local study sites and the central study database: Nikolaos Koutsouleris, Anne Ruef, Lisa Hahn, Dominic B. Dwyer, Shalaila Haas, Linda A. Antonucci, Alkomiet Hasan, Claudius Hoff, Ifrah Khanyaree, Aylin Melo, Susanna Muckenhuber-Sternbauer, Yanis Köhler, Ömer Öztürk, Nora Penzel, David Popovic, Adrian Rangnick, Sebastian von Saldern, Rachele Sanfelici, Moritz Spangemacher, Ana Tupac, Maria Fernanda Urquijo-Castro, Johanna Weiske, Antonia Wosgien, and Camilla Krämer (Department ofPsychiatry and Psychotherapy, Ludwig-Maximilian-University, Germany); Lana Kambeitz-Ilankovic, Joseph Kambeitz, Julian Wenzel, Stephan Ruhrmann, Karsten Blume, Dennis Hedderich, Dominika Julkowski, Nathalie Kaiser, Thorsten Lichtenstein, Ruth Milz, Alexandra Nikolaides, Tanja Pilgram, Mauro Seves, and Martina Wassen (Department of Psychiatry and Psychotherapy, University of Cologne, Germany); Christina Andreou, Stefan Borgwardt, André Schmidt, Anita Riecher-Rössler, Laura Egloff, Fabienne Harrisberger, Ulrike Heitz, Claudia Lenz, Letizia Leanza, Amatya Mackintosh, Renata Smieskova, Erich Studerus, Anna Walter, and Sonja Widmayer (Department of Psychiatry, Psychiatric University Hospital, University of Basel, Switzerland); Stephen J. Wood, Rachel Upthegrove, Paris Alexandros Lalousis, Chris Day, Sian Lowri Griffiths, Mariam Iqbal, Mirabel Pelton, Pavan Mallikarjun, Alexandra Stainton, and Ashleigh Lin (Institute for Mental Health and School of Psychology, University of Birmingham, UK); Jarmo Hietala, Raimo K.R. Salokangas, Alexander Denissoff, Anu Ellilä, Tiina From, Markus Heinimaa, Tuula Ilonen, Päivi Jalo, Heikki Laurikainen, Antti Luutonen, Akseli Mäkela, Janina Paju, Henri Pesonen, Reetta-Liina Säilä, Anna Toivonen, and Otto Turtonen (Department of Psychiatry, University of Turku, Finland); Frauke Schultze-Lutter, Eva Meisenzahl, Alexandra Korda, Susanne Neufang, Christian Schmidt-Kraepelin, Henrik Rohner, Sonja Botterweck, Norman Kluthausen, Gerald Antoch, Julian Caspers, and Hans-Jörg Wittsack (Department of Psychiatry, Psychiatric University Hospital LVR/Heinrich-Heine-University Düsseldorf, University of Düsseldorf, Germany); Pierluigi Selvaggi, Alessandro Bertolino, Giuseppe Blasi, Giulio Pergola, Grazia Caforio, Leonardo Fazio, Tiziana Quarto, Barbara Gelao, Raffaella Romano, Ileana Andriola, Andrea Falsetti, Marina Barone, Roberta Passiatore, and Marina Sangiuliano (Department of Basic Medical Science, Neuroscience and Sense Organs, University of Bari Aldo Moro, Italy); Rebekka Lencer, Marian Surmann, Olga Bienek, and Udo Dannlowski (Department of Psychiatry and Psychotherapy, University of Münster, Germany); Ana Beatriz Solana, Manuela Abraham, and Timo Schirmer (GE Global Research, Inc); Paolo Brambilla, Carlo Altamura, Marika Belleri, Francesca Bottinelli, Adele Ferro, and Marta Re (Department of Neuroscience and Mental Health, Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Workgroup of Paolo Brambilla, University of Milan, Italy); Emiliano Monzani and Maurizio Sberna (Programma 2000, Niguarda Hospital, Workgroup of Paolo Brambilla, University of Milan, Italy); Giampaolo Perna, Maria Nobile, and Alessandra Alciati (San Paolo Hospital, Workgroup of Paolo Brambilla, University of Milan, Italy); Armando D’Agostino and Lorenzo Del Fabro (Villa San Benedetto Menni, Albese con Cassano, Workgroup of Paolo Brambilla, University of Milan, Italy); Matteo Balestrieri, Carolina Bonivento, Giuseppe Cabras, and Franco Fabbro (Department of Medical Area, Workgroup of Paolo Brambilla, University of Udine, Italy); and Marco Garzitto and Sara Piccin (IRCCS Scientific Institute E. Medea, Polo FVG, Workgroup of Paolo Brambilla, University of Udine, Italy); Christos Pantelis (Melbourne Neuropsychiatry Centre, Department of Psychiatry, University of Melbourne \u0026amp; Melbourne Health, Australia), Christos Davatzikos (Department of Radiology, University of Pennsylvania School of Medicine, 3700 Hamilton Walk, Philadelphia, PA 19104).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSilverstein, S.M. and A. Lai, \u003cem\u003eThe phenomenology and neurobiology of visual distortions and hallucinations in schizophrenia: an update.\u003c/em\u003e Frontiers in Psychiatry, 2021. \u003cstrong\u003e12\u003c/strong\u003e: p. 684720.\u003c/li\u003e\n\u003cli\u003eButler, P.D., S.M. Silverstein, and S.C. Dakin, \u003cem\u003eVisual Perception and Its Impairment in Schizophrenia.\u003c/em\u003e Biological Psychiatry, 2008. \u003cstrong\u003e64\u003c/strong\u003e(1): p. 40-47.\u003c/li\u003e\n\u003cli\u003eKing, D.J., et al., \u003cem\u003eA review of abnormalities in the perception of visual illusions in schizophrenia.\u003c/em\u003e Psychonomic Bulletin \u0026amp; Review, 2017. \u003cstrong\u003e24\u003c/strong\u003e(3): p. 734-751.\u003c/li\u003e\n\u003cli\u003eSchultze-Lutter, F., et al., \u003cem\u003eEPA guidance on the early detection of clinical high risk states of psychoses.\u003c/em\u003e European Psychiatry, 2015. \u003cstrong\u003e30\u003c/strong\u003e(3): p. 405-416.\u003c/li\u003e\n\u003cli\u003eAddington, J., et al., \u003cem\u003eProgression from being at-risk to psychosis: next steps.\u003c/em\u003e npj Schizophrenia, 2020. \u003cstrong\u003e6\u003c/strong\u003e(1): p. 27.\u003c/li\u003e\n\u003cli\u003eSchultze-Lutter, F., et al., \u003cem\u003eRevisiting the Basic Symptom Concept: Toward Translating Risk Symptoms for Psychosis into Neurobiological Targets.\u003c/em\u003e Frontiers in Psychiatry, 2016. \u003cstrong\u003e7\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eSilverstein, S.M., \u003cem\u003eVisual Perception Disturbances in Schizophrenia: A Unified Model\u003c/em\u003e, in \u003cem\u003eThe Neuropsychopathology of Schizophrenia: Molecules, Brain Systems, Motivation, and Cognition\u003c/em\u003e, M. Li and W.D. Spaulding, Editors. 2016, Springer International Publishing: Cham. p. 77-132.\u003c/li\u003e\n\u003cli\u003eSilverstein, S.M., et al., \u003cem\u003ePerceptual organization, the disorganization syndrome, and context processing in chronic schizophrenia.\u003c/em\u003e Schizophrenia Research, 2000. \u003cstrong\u003e43\u003c/strong\u003e(1): p. 11-20.\u003c/li\u003e\n\u003cli\u003eCimmer, C., et al., \u003cem\u003eAbnormal neurological signs, visual contrast sensitivity, and the deficit syndrome of schizophrenia.\u003c/em\u003e Progress in Neuro-Psychopharmacology and Biological Psychiatry, 2006. \u003cstrong\u003e30\u003c/strong\u003e(7): p. 1225-1230.\u003c/li\u003e\n\u003cli\u003eAd\u0026aacute;mek, P., V. Langov\u0026aacute;, and J. Hor\u0026aacute;ček, \u003cem\u003eEarly-stage visual perception impairment in schizophrenia, bottom-up and back again.\u003c/em\u003e Schizophrenia, 2022. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 27.\u003c/li\u003e\n\u003cli\u003eKeane, B.P., et al., \u003cem\u003eFunctional dysconnectivity of visual and somatomotor networks yields a simple and robust biomarker for psychosis.\u003c/em\u003e Molecular Psychiatry, 2024.\u003c/li\u003e\n\u003cli\u003eHebart, M.N. and G. Hesselmann, \u003cem\u003eWhat visual information is processed in the human dorsal stream?\u003c/em\u003e Journal of Neuroscience, 2012. \u003cstrong\u003e32\u003c/strong\u003e(24): p. 8107-8109.\u003c/li\u003e\n\u003cli\u003eRamenghi, L.A., et al., \u003cem\u003eVisual performance and brain structures in the developing brain of pre-term infants.\u003c/em\u003e Early Human Development, 2010. \u003cstrong\u003e86\u003c/strong\u003e(1, Supplement): p. 73-75.\u003c/li\u003e\n\u003cli\u003eRomer, A.L., et al., \u003cem\u003eReplicability of structural brain alterations associated with general psychopathology: evidence from a population-representative birth cohort.\u003c/em\u003e Molecular psychiatry, 2021. \u003cstrong\u003e26\u003c/strong\u003e(8): p. 3839-3846.\u003c/li\u003e\n\u003cli\u003eElliott, M.L., et al., \u003cem\u003eA connectome-wide functional signature of transdiagnostic risk for mental illness.\u003c/em\u003e Biological psychiatry, 2018. \u003cstrong\u003e84\u003c/strong\u003e(6): p. 452-459.\u003c/li\u003e\n\u003cli\u003eSchwarzer, J.M., et al., \u003cem\u003eThe impact of visual dysfunctions in recent-onset psychosis and clinical high-risk state for psychosis.\u003c/em\u003e Neuropsychopharmacology, 2022. \u003cstrong\u003e47\u003c/strong\u003e(12): p. 2051-2060.\u003c/li\u003e\n\u003cli\u003eKorda, A.I., et al., \u003cem\u003eBrain texture as a marker of transdiagnostic clinical profiles in patients with recent-onset psychosis and depression.\u003c/em\u003e Nature Mental Health, 2024. \u003cstrong\u003e2\u003c/strong\u003e(1): p. 76-87.\u003c/li\u003e\n\u003cli\u003eKorda Alexandra I, et al., \u003cem\u003eIdentification of texture MRI brain abnormalities on first-episode psychosis and clinical high-risk subjects using explainable artificial intelligence.\u003c/em\u003e Translational Psychiatry, 2022. \u003cstrong\u003e12\u003c/strong\u003e(1): p. 481.\u003c/li\u003e\n\u003cli\u003eKorda, A.I., et al., \u003cem\u003eIdentification of voxel-based texture abnormalities as new biomarkers for schizophrenia and major depressive patients using layer-wise relevance propagation on deep learning decisions.\u003c/em\u003e Psychiatry Research: Neuroimaging, 2021. \u003cstrong\u003e313\u003c/strong\u003e: p. 111303.\u003c/li\u003e\n\u003cli\u003eHu, Y., et al., \u003cem\u003eBrain Fractal Dimension and Machine Learning can predict first-episode psychosis and risk for transition to psychosis.\u003c/em\u003e Computers in Biology and Medicine, 2025. \u003cstrong\u003e193\u003c/strong\u003e: p. 110333.\u003c/li\u003e\n\u003cli\u003eLalousis, P.A., et al., \u003cem\u003eHeterogeneity and Classification of Recent Onset Psychosis and Depression: A Multimodal Machine Learning Approach.\u003c/em\u003e Schizophrenia Bulletin, 2021. \u003cstrong\u003e47\u003c/strong\u003e(4): p. 1130-1140.\u003c/li\u003e\n\u003cli\u003eKoutsouleris, N., et al., \u003cem\u003ePrediction Models of Functional Outcomes for Individuals in the Clinical High-Risk State for Psychosis or With Recent-Onset Depression: A Multimodal, Multisite Machine Learning Analysis.\u003c/em\u003e JAMA Psychiatry, 2018. \u003cstrong\u003e75\u003c/strong\u003e(11): p. 1156-1172.\u003c/li\u003e\n\u003cli\u003eKay, S.R., A. Fiszbein, and L.A. Opler, \u003cem\u003eThe Positive and Negative Syndrome Scale (PANSS) for Schizophrenia.\u003c/em\u003e Schizophrenia Bulletin, 1987. \u003cstrong\u003e13\u003c/strong\u003e(2): p. 261-276.\u003c/li\u003e\n\u003cli\u003eAndreasen, N.C., \u003cem\u003eThe Scale for the Assessment of Negative Symptoms (SANS): Conceptual and Theoretical Foundations.\u003c/em\u003e British Journal of Psychiatry, 1989. \u003cstrong\u003e155\u003c/strong\u003e(S7): p. 49-52.\u003c/li\u003e\n\u003cli\u003eBeck AT, S.R., Brown GK., \u003cem\u003eBeck Depression Inventory-II\u003c/em\u003e. 1996: San Antonio, TX: Psychological Association.\u003c/li\u003e\n\u003cli\u003eSchultze-Lutter, F., et al., \u003cem\u003eSchizophrenia proneness instrument, adult version (SPI-A).\u003c/em\u003e Rome: Giovanni Fioriti, 2007.\u003c/li\u003e\n\u003cli\u003eKorda, A.I., et al., \u003cem\u003eChaos analysis of the cortical boundary for the recognition of psychosis.\u003c/em\u003e Journal of Psychiatry and Neuroscience, 2023. \u003cstrong\u003e48\u003c/strong\u003e(2): p. E135.\u003c/li\u003e\n\u003cli\u003eDiamond, A., S.M. Silverstein, and B.P. Keane, \u003cem\u003eVisual system assessment for predicting a transition to psychosis.\u003c/em\u003e Translational Psychiatry, 2022. \u003cstrong\u003e12\u003c/strong\u003e(1): p. 351.\u003c/li\u003e\n\u003cli\u003ePodwalski, P., et al. \u003cem\u003ePsychopathology and Integrity of the Superior Longitudinal Fasciculus in Deficit and Nondeficit Schizophrenia\u003c/em\u003e. Brain Sciences, 2022. \u003cstrong\u003e12\u003c/strong\u003e, DOI: 10.3390/brainsci12020267.\u003c/li\u003e\n\u003cli\u003eKim, S.-H. and J.-S. Kim, \u003cem\u003eEye movement abnormalities in middle cerebellar peduncle strokes.\u003c/em\u003e Acta Neurologica Belgica, 2019. \u003cstrong\u003e119\u003c/strong\u003e(1): p. 37-45.\u003c/li\u003e\n\u003cli\u003eAlonso-S\u0026aacute;nchez, M.F., et al., \u003cem\u003ePerplexity of utterances in untreated first-episode psychosis: an ultra\u0026ndash;high field MRI dynamic causal modelling study of the semantic network.\u003c/em\u003e Journal of Psychiatry and Neuroscience, 2024. \u003cstrong\u003e49\u003c/strong\u003e(4): p. E252.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"translational-psychiatry","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"tp","sideBox":"Learn more about [Translational Psychiatry](http://www.nature.com/tp/)","snPcode":"41398","submissionUrl":"https://mts-tp.nature.com/cgi-bin/main.plex","title":"Translational Psychiatry","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"recent onset psychosis, recent onset depression, clusters of brain texture relevance, explainable artificial intelligence","lastPublishedDoi":"10.21203/rs.3.rs-7178786/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7178786/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eDeeper understanding of Subtle Visual Dysfunctions (VisDys) in the early stage of mental illness and their neurobiological underpinnings, as reflected by microstructural brain texture features, could advance our understanding of the underlying disease perceptual mechanisms that mediate susceptibility to psychosis. In this study, we aim a) to investigate the utility of brain texture features for the prediction of VisDys in recent onset psychosis (ROP) and clinical high-risk syndromes for psychosis (CHR-P), respectively, b) to test prediction models established in ROP and CHR-P in an independent validation sample with recent onset depression (ROD) diagnoses and c) to test for symptom expression related brain features associated with VisDys. sMRI were acquired in a training sample including 128 ROP (67 patients with VisDys), 134 CHR-P (71 patients with VisDys). Independent validation sets included 46 ROP (19 with VisDys), 124 CHR-P (68 patients with VisDys) and a sample of 256 ROD (50 patients with VisDys). Due to the heterogeneity of the CHR-P group, a clustering analysis was performed to identify proof of concept clusters of brain relevance specifically associated with symptoms in patients with and without VisDys, respectively. Both classification schemas in ROP and CHR-P presented good prediction accuracy in the independent validation samples of ROP, CHR-P and ROD, respectively.\u003c/p\u003e","manuscriptTitle":"Brain texture alterations predict subtle visual perceptual dysfunctions in recent onset psychosis and clinical high-risk state","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-29 08:17:38","doi":"10.21203/rs.3.rs-7178786/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2025-09-08T09:14:10+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2025-08-12T06:44:38+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2025-08-03T14:30:03+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2025-07-28T06:51:43+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2025-07-24T16:37:25+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2025-07-24T16:27:54+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-22T12:54:11+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-22T12:53:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Translational Psychiatry","date":"2025-07-21T14:51:07+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"translational-psychiatry","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"tp","sideBox":"Learn more about [Translational Psychiatry](http://www.nature.com/tp/)","snPcode":"41398","submissionUrl":"https://mts-tp.nature.com/cgi-bin/main.plex","title":"Translational Psychiatry","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e9c031f3-94a9-41be-9454-7c7fe2c1de11","owner":[],"postedDate":"July 29th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":52076817,"name":"Health sciences/Diseases/Psychiatric disorders/Schizophrenia"},{"id":52076818,"name":"Health sciences/Diseases/Psychiatric disorders/Depression"}],"tags":[],"updatedAt":"2026-02-21T08:10:10+00:00","versionOfRecord":{"articleIdentity":"rs-7178786","link":"https://doi.org/10.1038/s41398-026-03840-x","journal":{"identity":"translational-psychiatry","isVorOnly":false,"title":"Translational Psychiatry"},"publishedOn":"2026-02-12 05:00:00","publishedOnDateReadable":"February 12th, 2026"},"versionCreatedAt":"2025-07-29 08:17:38","video":"","vorDoi":"10.1038/s41398-026-03840-x","vorDoiUrl":"https://doi.org/10.1038/s41398-026-03840-x","workflowStages":[]},"version":"v1","identity":"rs-7178786","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7178786","identity":"rs-7178786","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}