Longitudinal Surface-Based Morphometry Changes in the Hippocampus in Dementia

Longitudinal Surface-Based Morphometry Changes in the Hippocampus in Dementia | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (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];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search Longitudinal Surface-Based Morphometry Changes in the Hippocampus in Dementia View ORCID Profile Salah Aziz , View ORCID Profile Romeo Penheiro , View ORCID Profile Cassandra Morrison , View ORCID Profile Peter Zhukovsky , View ORCID Profile John AE Anderson , the Alzheimer’s Disease Neuroimaging Initiative (ADNI) doi: https://doi.org/10.1101/2025.04.07.25325336 Salah Aziz 1 Department of Cognitive Science, Carleton University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Salah Aziz Romeo Penheiro 2 Cognitive Neuropsychology of Daily Life (CNDL) Laboratory, University of Houston Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Romeo Penheiro Cassandra Morrison 3 Department of Psychology, Carleton University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Cassandra Morrison Peter Zhukovsky 4 Brain Health Imaging Center, Center for Addiction and Mental Health , Toronto, ON Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Peter Zhukovsky John AE Anderson 1 Department of Cognitive Science, Carleton University 3 Department of Psychology, Carleton University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for John AE Anderson For correspondence: johnanderson3{at}cunet.carleton.ca Abstract Full Text Info/History Metrics Supplementary material Data/Code Preview PDF Abstract The hippocampus is central to Alzheimer’s disease (AD), characterized by atrophy and cognitive decline. While volume loss is well-documented, surface-based morphometric (SBM) features—curvature, gyrification, and thickness—remain less explored. Using T1-weighted MRI data from the Alzheimer’s Disease Neuroimaging Initiative (4,617 timepoints; CN: 475, MCI: 673, AD: 269), hippocampal subfields were analyzed with HippUnfold. Linear mixed effects models examined volume and SBM changes, tracking cognitive trajectories in stable (n = 1017) and progressing (n = 301) individuals. Focusing on CA1, the stable AD group showed significant volume reductions (β = −1.01, p < .001) compared to CN (β = 6.19, p < .001). SBM metrics significantly increased over time across subregions (e.g., curvature: CN: β = −0.72, p < .001; AD: β = 0.006, p < .001), though gyrification did not reach significance in bilateral CA1, CA3, CA4, and subiculum. Time-dependent interactions indicated progressive volume loss and SBM increases across groups (all p’s < .001). Notably, SBM metrics predicted cognitive improvements in AD. While volume loss remains a key AD marker, it may not capture early morphometric changes. SBM features provide additional insights, underscoring the need for integrated volumetric and surface-based analyses to refine disease detection and therapeutic strategies. 1 Introduction Progressive hippocampal atrophy is a hallmark of Alzheimer’s disease (AD), crucial for identifying neurodegeneration in AD and its prodromal stage, mild cognitive impairment (MCI). Hippocampal atrophy is linked to memory decline, making it a key focus in AD studies. Most studies of brain structure in AD have concentrated on volumetric changes within the medial temporal lobe. However, surface-based metrics like curvature, gyrification, and thickness remain largely unexplored due to challenges in hippocampal segmentation on routine MRI scans ( Insausti & Amaral, 2004 ; Chang et al., 2018 ; Wisse et al., 2017 ; Yushkevich et al., 2015 ). Although post-mortem studies reveal variability in hippocampal subfields constrained by the folded structure ( Duvernoy, 1998 ), segmentation protocols remain challenging, despite harmonization efforts (La Joie et al., 2020; Olsen et al., 2019 ; Wisse et al., 2017 ). Recent advances in automated segmentation, particularly deep learning with convolutional neural networks, offer faster, more accurate methods than manual segmentation ( Goubran et al., 2020 ; Henschel et al., 2020 ; Thyreau et al., 2018 ). Tools like HippUnfold ( Dekraker et al., 2021 , 2022 ) enable the extraction of SBM metrics, providing insights beyond volume into surface features sensitive to early neurodegeneration. HippUnfold reconstructs a model of hippocampal tissue, enabling precise subfield parcellation without distortions and capturing metrics like curvature, thickness, and gyrification—each associated with AD-relevant structural changes ( Dale et al., 1999 ; Fischl & Dale, 2000 ; Schaer et al., 2008 ). Additional metrics beyond volume include methods like Scoring by Nonlocal Image Patch Estimator (SNIPE) grading. SNIPE, a non-local patch-based measure of anatomical similarity combined with hippocampal segmentation, has demonstrated enhanced sensitivity over traditional volume-based measures in detecting group differences across the spectrum of cognitive decline ( Coupé et al., 2012 ) Recent work using SNIPE grading revealed significantly higher classification accuracies for distinguishing between cohorts, outperforming volume-based approaches. ( Morrison et al., 2023a ; 2023b ) These findings suggest that SNIPE grading is not only more sensitive to group differences but also provides stronger associations with cognitive trajectories and disease states than hippocampal volume alone. Building on this, our study takes a complementary approach by incorporating hippocampal morphometry features such as curvature, gyrification, and thickness in addition to volume. By doing so, we aim to further elucidate the structural contributions of the hippocampus to cognition and its sensitivity to group differences across the cognitive decline spectrum. Our study focuses on hippocampal subfields implicated in AD: CA1-CA4, dentate gyrus (DG), subiculum (Sub), and subiculum-reticular layer of the medial temporal lobe (SRLM). Typically, CA1, DG, and Sub show atrophy in AD, whereas CA2 and CA3 are often spared ( Christopher-Hayes et al., 2023 ). These subfields correlate with cognitive functions impaired in AD, such as episodic memory ( Tulving, 1972 ; Eichenbaum, 2017 ), with CA1, CA4, and DG volume reductions associated with memory decline in MCI and mild AD ( Broadhouse et al., 2019 ; Kerchner et al., 2012). Other cognitive domains, such as visuospatial processing, executive functioning, and language, are also impacted ( Rizzo et al., 2000 ; Jacobs et al., 2015 ; Manning et al., 2023). Using the ADNI dataset, we aim to investigate hippocampal subfield changes in individuals classified as CN, MCI, and AD, as this has not been extensively studied using this method. While previous research has often focused on cross-sectional analyses of hippocampal subfields, our study takes a novel longitudinal approach, examining changes across four ADNI phases (ADNI 1, GO, 2, 3) and covering a range of cognitive stages. We hypothesized that individuals with AD would exhibit significant hippocampal atrophy compared to healthy controls, with pronounced atrophy in the CA1 and subiculum subregions. We further anticipated that these changes would provide unique insights into the progression of cognitive decline over time. Due to the novelty of the metric device, we had limited a priori hypotheses regarding the direction of surface metrics, aside from expecting decreases in cortical thickness. While HippUnfold has mainly been applied cross-sectionally, our study extends its use longitudinally within the ADNI dataset, enabling the tracking of hippocampal metric changes over time and deepening our understanding of AD progression. 2 Methods 2.1 ADNI Data used in the preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). The ADNI was launched in 2003 as a public-private partnership, led by Principal Investigator Michael W. Weiner, MD. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), positron emission tomography (PET), other biological markers, and clinical and neuropsychological assessment can be combined to measure the progression of mild cognitive impairment (MCI) and early Alzheimer’s disease (AD) (for more information, visit http://www.adni-info.org ). The study received ethical approval from the review boards of all participating institutions. Written informed consent was obtained from participants or their study partner. 2.2 ADNI Study Data MRI metadata, neuropsychological data, and demographic data were obtained on ADNI’s public data website, and merged with the HippUnfold metric data. We used rolling joins to eliminate scans from the study in which the relevant neuropsychological batteries were greater than 180 days from the MRI scan, as the next sequence of scans were scheduled 6 months apart from each other ( Morrison et al. 2023a ). We obtained demographic data and cognitive scores from the ADNI study data portal. On the recommendation of the Center for Psychometric Analysis of Aging and Neurodegeneration (CPAAN) at University of Washington for ADNI, we used the data harmonized from the Alzheimer’s Disease Sequencing Project Harmonization Consortium (ADSP-PHC), an attempt to harmonize cognitive domain scores across multiple studies of aging ( Mukherjee et al., 2023 ). Cognitive composite scores are divided into four domains: memory (PHC_MEM), executive functioning (PHC_EXF), visuospatial processing (PHC_VSP), and language processing (PHC_LAN). The composite scores were standardized using Z-standardization. The analysis utilized data from the ADNI study, which included some missing neuropsychological measures across participants. This missing data is noted in the analysis. 2.3 Participants In this study, we emphasize the significance of participant retention throughout the entire duration of the longitudinal study. By maintaining a consistent cohort of participants in our data selection, we would be able to gain a more comprehensive understanding of the progression of cognitive decline, enabling us to capture the genuine, dynamic nature of its changes, as well as to ensure power and internal validity in our study ( Fewtrell et al., 2008 ; Siddiqi et al., 2008 ; Gupta et al., 2015 ). In general, eligibility criteria for ADNI include ages 55-90, having a baseline diagnosis of either CN, MCI or early-stage AD, a minimum of six years of education, a sufficient command of English to ensure they can adequately complete the study assessments, and being able to undergo MRI scanning procedures, exclusions include scans within one year of enrollment showing infection, infarction, or focal lesions; bodily augmentations or foreign objects interfering with MRI; and significant systemic illnesses or unstable medical conditions that hinder protocol compliance. Other inclusion and exclusion criteria can be found at https://adni.loni.usc.edu/help-faqs/adni-documentation/ . To ensure the consistency and reliability of our findings, we excluded participants whose diagnostic status regressed at any point in their time during the ADNI study, as it would bring uncertainty as to the interpretation of cognitive trajectories. Attrition rates in the ADNI cohorts are particularly high ( Insel et al., 2019 ), and for this reason, we chose to select participants that had at least two consecutive scans throughout their involvement. Although here we reference the AD diagnosis as such, our identification of diagnosis of participants with Alzheimer’s disease is based on a clinical diagnosis in ADNI, and not one founded on neurobiological identification of tau and amyloid deposits. A total of 1,429 participants were initially included in this study. After excluding 27 participants due to segmentation failures, pre-processing issues, or visible artifacts in scans, and an additional 87 participants due to fluctuations in diagnosis, the final number of participants analysed was N = 1,315, with an average of 3.51 annual scans (4,617 timepoints). Populations of diagnostic groups include cognitively-normal (CN): 475; mild-cognitive impairment (MCI): 673; Alzheimer’s disease (AD): 269. To better identify the trajectory of cognitive impairment of the participants, we categorized the participants by the clinical diagnosis measured at their final timepoint in contrast to their baseline diagnostic group. We made the following categorical scheme of trajectories: stable , or non-progressors , (n = 1017): (CN, n = 411; MCI, n = 366; AD, n = 240) and progressor s (n =301): (CN to MCI, n = 51; CN to AD, n =23; MCI to AD, n = 227). We chose to group together the CN progressors due to a small sample size. View this table: View inline View popup Download powerpoint Table 1 Demographic Information by Group 2.4 Structural MRI Acquisition and HippUnfold Processing Participants underwent scanning using 1.5T and 3.0T T1-weighted protocols. Full criteria and details available at http://www.adni-info.org . Original raw T1-weighted scans were retrieved from the ADNI database and processed using HippUnfold v.1.1 on CBRAIN ( Sherif et al., 2014 ). The hippocampal data were first intensity-corrected with the N4 algorithm from the Advanced Normalization Toolkit (ANTs) ( Avants et al., 2008 ). Next, images were aligned to the CITI168 atlas ( Pauli et al., 2018 ) through affine registration performed with NiftyReg ( Modat et al., 2010 ). This alignment was refined by combining it with a precomputed transform from the CITI168 atlas to an oblique orientation along the hippocampus’s long axis, using Convert3D ( Yushkevich et al., 2006 ). After alignment, the images were resampled to a voxel size of 0.3 mm³ and cropped to dimensions of 128 × 256 × 128 voxels, centering on the left and right hippocampi of the CITI168 atlas. To ensure consistency, left hippocampi were sagittally flipped to match the orientation of right hippocampi. HippUnfold metrics were Z-standardized to facilitate comparison. Volumetric measurements were available for SRLM and Cyst, while thickness metrics were not generated for DG. 2.5 Residuals We controlled for estimated-total intracranial volume (eTIV), age, sex, education level, and scanner-site. eTIV was calculated by utilizing Fastsurfer v.2.12., a segmentation and volume pipeline ( Faber et al., 2022 ; Henschel et al., 2020 ; Henschel et al., 2022 ). Parcellation of eTIV via total white and grey matter was obtained by using the Desikan-Killianny-Tourville (DKT) (Klein & Tourville, 2012) protocol generated by Fastsurfer; the volume metric of the parcellation was then calculated using FSLstats in FSL (Woolrich et al., 2009). 2.6 Statistical Analyses We first conducted statistical analyses to explore the differences between groups. To determine if the observed differences in age, education, cognitive scores, and study followup-years among the diagnostic groups were statistically significant, we performed a series of one-way ANOVAs. We examined biological sex separately as a Chi-Square test as a categorical variable. Statistical analysis was conducted in “R” v.4.3.2 ( R Core Team, 2021 ). We used a linear mixed effects model with the package lme4 to investigate the hierarchical mixed effects of hippocampal morphometry in participants over time. Due to convergence and singularity issues--likely stemming from the complexity of the nested response structure and the volume of data--we chose to focus on specific combinations of hemisphere, hippocampal subfield, and anatomical metric, rather than attempting a single comprehensive model. For each combination, a separate model was implemented using the data subset accordingly: Residualized Hippocampus Measure ∼ Progression Group * Followup Years + Covariates (1 + Followup Years | Participant ID) By using this subsetting strategy, we ensured reliable parameter estimates, allowing us to draw meaningful conclusions regarding the effects of hippocampal morphometry on cognitive scores. This approach resulted in 50 distinct models being implemented, corrected for multiple comparisons. A trajectory plot ( Figure 1 ) was generated to visualize changes over time for each hemisphere, hippocampal label, and metric (volume, curvature, gyrification, thickness). Data was grouped by diagnosis and linear models with the interaction term diagnosis and follow-up years were fitted to estimate slopes of over time. Post-hoc pairwise comparisons of slopes were conducted using the emmeans package, with corrections of multiple comparisons with the False Discovery Rate (FDR) approach using the p.adjust function in R, corresponding to the Benjamini-Hochberg procedure. ( Benjamini & Hochberg, 1995 ) Significant results were noted in the analysis. Download figure Open in new tab Figure 1: Longitudinal Change in Left CA1 Morphometry for Different Diagnostic Groups Note : Residualized surface-based morphometry (SBM) change over time across four metrics: (A) Volume, (B) Curvature, (C) Gyrification, and (D) Thickness. Trajectories for five progressive groups—Stable CN, Stable MCI, Stable AD, Progressive CN to MCI/AD, and Progressive MCI to AD—are shown. Solid lines represent best-fit slopes, while shaded paths depict individual trajectories. Slope values, reported to four decimal places for clarity, may appear identical despite underlying differences. Pairwise comparisons reveal significant differences in volume across all groups (p < 0.001), and between CN and AD, MCI and AD, and transitions from CN to MCI/AD and MCI/AD for other metrics (p < 0.001). View this table: View inline View popup Download powerpoint Table 2: Breakdown of Hippocampal Subfield Models by Anatomical Measure and Hemisphere The diagnostic group Stable CN was set as the reference level for the intercept and group level effects. This model also provided us with participant level variance of slopes and interaction with timepoints. The model was estimated using Restricted Maximum Likelihood (REML), a method that provides accurate estimates of variance components by accounting for the fixed effects, along with the Bound Optimization By Quadratic Approximation (BOBYQA) optimizer, which ensures the model efficiently converges even in complex arrangements of data where traditional methods may struggle. This approach was used to predict standardized and residualized modality values with diagnostic groups and follow-up years. Then we looked at how random effects from the models accounted for individual variability. For the time-dependent interactions, we transition to visual representations of among diagnosis groups, focusing on how the average slopes representing changes over time are spatially distributed across various hippocampal regions using Connectome Workbench ( Van Essen & The Human Connectome Project, 2012 ). We assess the interaction effects between Follow-up Years and the diagnosis groups. These interactions reveal how the slopes for different diagnosis groups diverge, highlighting the temporal dynamics associated with the progression of diagnosis. Afterwards, we ran linear models to assess how the metrics of morphometry can predict changes in cognitive composite scores. Afterwards we ran sensitivity analyses due to differences of MRI tesla strengths and high variability in our PCA analysis. To control for multiple comparisons, we applied the FDR adjustment 6 . To predict cognitive scores, we fitted linear regression models for each particular hemisphere, metric correspondant to a hippocampal subfield with the slope and progressive change group as predictors for each cognitive domain. The general formula for each regression model was: Cognitive Score ∼ Morphometry * Diagnostic Group. 3 Results 3.1 Descriptive Statistics In addition to the one-way ANOVAs conducted for continuous variables, a chi-square test was performed to examine the association between diagnostic group and sex. The results showed a significant association between sex and diagnostic group, χ²(4, N =1318) = [2158.3], [p < .001]. 7 3.2 Main Findings The analysis of hippocampal subregions revealed several broad trends across different metrics. Figure 1 illustrates a general pattern where decreases in volume are often accompanied by increases in SBM metrics. This relationship is particularly evident in individuals diagnosed with Stable AD and progressive groups when compared to the Stable CN controls. This pattern of decreased volume but increased curvature, gyrification and thickness are evident in bilateral CA1, Sub, and the DG. Interestingly, a distinct pattern emerged in the volume and SBM metrics across hippocampal subregions. While volumetric measures in bilateral CA2 and CA3 regions remained relatively stable across all cohorts, SBM metrics such as curvature, gyrification, and thickness showed consistent increases , suggesting shifts in surface topology despite stable volume. A similar trend was observed in the CA4 subregion, where slight volumetric decreases were accompanied by upward trends in other SBM metrics. These findings highlight the complementary insights provided by SBM metrics, particularly in capturing morphological changes that volume alone may not detect. Moreover, divergent trends were noted in the left Sub, where a decrease in volume was observed alongside a relatively flat curvature slope. These patterns underscore the heterogeneous nature of hippocampal subfield changes and the importance of integrating volumetric and SBM metrics to comprehensively assess structural changes associated with cognitive decline. While the visual patterns suggest overlapping trajectories, the results from our linear mixed effect models provide additional insight into these changes. Specifically, the models reveal significant differences in the effect sizes between the diagnostic groups. In particular, the comparison between Stable CN and Stable AD groups shows clear distinctions in the rates of structural change. 3.3 Fixed Effects Table 3 presents the fixed effects of the diagnostic group on hippocampal metrics across subregions, reporting findings that remained significant after false discovery rate (FDR) correction. While all comparisons were subjected to FDR adjustment, the effects described here reflect the most robust surviving results. Individuals with stable AD exhibited significant volume reductions and increases in SBM metrics compared to CN individuals. The largest volume reduction was observed in L CA1 (β = −1.01, p < .001), followed by SRLM and Sub, indicating substantial atrophy in these regions. Curvature showed the largest effect sizes for SBM metrics, ranging up to .006 for CA1 curvature for Stable AD (p < 0.01). In contrast, gyrification had the smallest estimates among the SBM metrics, and insignificant in both Stable MCI and AD across CA1, CA3, CA4, and the DG. View this table: View inline View popup Table 3: Summary of One-Way ANOVA Results Across Demographic and Cognitive Variables by Diagnostic Group View this table: View inline View popup Table 4: Fixed Effects of Diagnostic Status on Hippocampal Metrics Across Subregions The covariates included in the models were significant for most metrics, with exceptions primarily in volume and L Sub thickness. Age was a significant predictor except for volume in L CA3 and L CA4. Education showed significance across all models except for volume in bilateral Cysts. Scanner site had no significant effects on any volume metric or L Sub thickness. Sex was not significant for volume in L CA2, bilateral CA4, and L DG. The random intercept standard deviation ranged from 0.60 to 0.03, with volume measures showing the highest variances and gyrification exhibiting the lowest. Random slope standard deviation variances ranged from 0.04 to 0.004, reflecting a similar trend, with volume metrics exhibiting higher variability. The correlation between the intercept and slope ranged from 0.04 to 0.64, with volume metrics showing the strongest correlation and curvature and gyrification exhibiting the weakest. Residual standard deviations ranged from −0.4 to −5.70E-5, with volume metrics displaying the highest residual variance and curvature metrics showing the lowest. 3.4 Slopes of Main Effects and Interaction Effects Table 5 indicates the significance of interactions retained by multiple comparisons, with the CA1 region serving as an exemplar, as the pattern of effects of the interaction terms are consistent with other significant interactions; see supplementary materials for full data. The trend of effects of volume decreasing and SBM metrics rising are consistent with the fixed effects in the models. For the volume metrics, reductions in volume had greatest negative effects for the MCI to AD group. For example, CA1 was significant in both the Stable AD group (β = −0.05) and in the transition from CN to MCI/AD (β = −0.03), yet MCI to AD showed (β = −0.06). However, significant increases were observed in CA2 for Stable AD (β = 0.04), and in CA3 for MCI to AD (β = .03). Reductions in the Sub, DG, SRLM, and Cyst regions were found in the comparisons between CN and MCI/AD, as well as in the MCI to AD group. SBM metrics showcased higher increases for the MCI to AD group in comparison to the Stable AD group. Whereas gyrification was not significant in the main effects for Stable AD in a number of subfields, the interaction for time shows significance. View this table: View inline View popup Table 5: Interactions of Follow-up Years and Diagnostic Group Surviving Multiple Comparisons Figure 2 illustrates the average slopes within each metric across diagnostic groups. For volume, CA1 exhibits the lowest slopes, while CA2 shows the highest. Curvature varies by hemisphere, with DG having the highest slopes in the left hemisphere but among the lowest in the right, while CA4 shows the opposite pattern. For gyrification, DG has the lowest slopes, whereas CA2 has the highest. In thickness, Sub displays some of the highest slopes, while CA4 has the lowest. Download figure Open in new tab Figure 2: Average Slopes of Morphometry by Diagnostic Group Note . (A)–(D) display left dorsal and unfolded hippocampal maps from Connectome Workbench, using label files with RGB values based on averaged slopes by metric, hemisphere, and diagnostic group, ordered by slope severity. Colors represent relative slope differences, with blue, white, and red indicating the lowest, median, and highest averages. See supplementary section for bilateral slope comparison. 3.5 Effects on Cognition We applied FDR adjustment for multiple comparisons to control for Type I errors. Overall, only 22.8% of results for interactions survived comparisons for significance. Our analysis revealed that among the models that reached statistical significance, memory is the predominant cognitive domain for the metrics that served as predictors, with visuospatial processing being the lowest. Figure 3 illustrates the range of effect sizes from these interactions. Interestingly, while volume was observed to have minimal effect sizes on cognition, the positive SBM metrics predicted positive effect sizes on cognition for the Stable AD group. Download figure Open in new tab Figure 3: Effect Sizes for Interactions of Morphometry with Diagnosis Note. This point-range plot of the left CA1 presents the interaction effects of diagnosis and slope on cognitive domains, executive functioning (EXF), language processing (LAN), and memory (MEM). Each panel corresponds to a specific subfield for the left hemisphere, with the point-ranges indicating the magnitude of the effects and standard error. 3.6 Sensitivity Analyses We found more significant findings in the 3.0T than the 1.5T data. In the 3.0T data, more significant findings were identified for curvature, gyrification, and thickness, whereas the 1.5T data showed more significant effects on volumetric measures. A subsequent analyses of 1.5 and 3.0T data collectively showed that there is a general trend that a majority of the findings share significant findings that are reported in our main analysis. However, there are some divergences. Specifically, the 1.5T data alone picked up decreases in effect sizes of CA1 volumes for time interactions from CN to MCI/AD, and Stable MCI; whereas for volume for the CA2 and CA3 regions there were increases of effect sizes than the 3.0 scans. 4 Discussion Our analyses have uncovered several noteworthy findings regarding hippocampal and cortical metrics in AD. As hypothesized, we observed greater negative effect sizes for volume reductions in AD, particularly in the CA1 and subiculum subregions. This is consistent with prior research and reinforces the role of these regions as biomarkers in AD progression. These findings align with Mak et al. (2016) , who reported significant atrophy in this area, and corroborates earlier studies by Apostolova et al. (2006) , La Joie et al. (2013) , and Zhao et al. (2019) , which highlighted the CA1 and subiculum as critical sites of neurodegeneration in AD. These findings reinforce the importance of these regions as biomarkers of AD progression. However, contrary to our expectation, the decrease of cortical thickness was not confirmed. Instead of the anticipated cortical thinning in AD, we observed an increase in cortical thickness. This result parallels findings from Phan et al. (2024) , who noted an anticorrelation between cortical thickness and brain atrophy, particularly in the anterior cingulate cortex (ACC). Possible explanations for this phenomenon include neuroplasticity and neuroinflammation. The increased cortical thickness could reflect adaptive neuroplastic changes or inflammatory responses altering cortical architecture. Another study utilizing HippUnfold examining age-related decline in thickness and surface area in the hippocampus reveals that there are patterns of decline in cortical thinning that are mostly linear, but are highly heterogeneous ( Yu et al, 2024 ). Cortical thickening has also been reported in early preclinical AD ( Montal et al., 2018 ), as well as in normal tau protein accumulation in AD cohorts ( Hojjati et al., 2023 ). We did not hypothesize an increase in gyrification that we observed in AD. This finding is reminiscent of recent work by Núñez et al. (2020) , which found increased gyrification in the entorhinal cortex in AD. This unexpected finding may be indicative of a compensatory response to white matter (WM) atrophy or neurogenesis in specific cortical areas. However, this hypothesis remains controversial, and further research is required to better understand the underlying mechanisms. Similarly, we also did not hypothesize an increase in cortical curvature, yet the increase in cortical curvature we documented may be a necessary correlate of the increased gyrification. This suggests a complex relationship between cortical folding and overall brain morphology in AD. The interplay between gyrification and curvature is an area that warrants further investigation to understand the adaptive or maladaptive changes occurring in AD. One possible explanation is the phenomenon of white matter pulling away from grey matter. As WM atrophy occurs in AD, the reduced volume of white matter may lead to a “stretching” effect on the overlying cortical gray matter. This could increase gyrification as the cortical surface is stretched in response to the shrinking underlying white matter. In summary, while the volume reductions in the HC align with expected patterns of neurodegeneration in AD, the observed increases in gyrification, cortical thickness, and curvature present novel insights into the disease’s impact on brain morphology, or to be more precise, morphology in the HC. These findings suggest both adaptive and pathological changes and highlight the need for further research to elucidate their mechanisms and implications for AD understanding and diagnosis. Our sensitivity analysis replicates previous findings that 3.0T is more sensitive to effect sizes and significant differences in neurodegeneration ( Chow et al., 2015 ; Leandrou et al., 2020). Although general trends are found within both field strengths, mixing data of these strengths have been shown to not drastically affect bias or variation between participants ( Ho et al., 2010 ). 5 Limitations While our findings are substantive, there are several limitations. First, the ADNI dataset lacks full harmonization to account for neuropsychiatric differences across its phases. Additionally, HippUnfold’s subfield definitions, based on 3DBigBrain ( DeKraker et al., 2020 ), differ slightly from traditional segmentation boundaries. For example, recent work by Hickling et al. (2024) shows CA4 associated with DG, and the Sub region divided into two substrates. Another limitation is the time gap between MRI scans and neuropsychological assessments; despite allowing a 180-day cutoff, closer tracking would provide a more accurate link between brain states and neuropsychiatric measures. Though HippUnfold represents an advanced deep-learning tool for hippocampal subfield segmentation, it has limitations, particularly regarding validation against manual segmentation. Similar to concerns raised with FreeSurfer, HippUnfold’s subfield classification may be less precise without high-resolution T2 data, and it has shown a tendency to overestimate subfield volumes ( Sghirripa et al., 2024 ). To mitigate volume measurement inaccuracies, we recommend including additional surface parameters in future analyses. 6 Conclusion In summary, our findings illuminate the complex interplay between structural changes in the HC and cognitive decline in AD. We discovered that both Stable CN and AD groups exhibited significant reductions in hippocampal volume, yet paradoxically, they demonstrated increased SBM effects, suggesting a nuanced adaptation in the brain’s architecture as it grapples with the onset of dementia. The significance of time interactions revealed that individuals transitioning from CN to MCI/AD and MCI to AD displayed particularly pronounced effects, reinforcing the notion that early interventions could be crucial in mitigating the trajectory of cognitive decline. Notably, language and executive functioning emerged as a powerful predictor in our linear regression analyses for SBM metrics, eclipsing the influence of volume, although the association between volume loss and memory decline in AD remained strong and undeniable. These results not only deepen our understanding of the morphological transformations occurring in the hippocampus as AD progresses but also underscore the necessity of examining surface-based parameters alongside traditional volumetric measures. As we look to the future, our work paves the way for broader investigations into the intricate relationships between hippocampal morphology, cognitive functions, and emotional health, offering hope for more comprehensive approaches to diagnosing and treating cognitive decline in AD. Data Availability All data produced in the present study are available upon reasonable request to the authors. http://www.loni.ucla.edu/ADNI https://github.com/salahaziz94/ADNI_HippUnfold Author Note We have no known conflict of interest to disclose. This project was supported by an NSERC Discovery Grant (DGECR-2022-00309) and a Canada Research Chair (Tier II, CRC-2020-00174) to JAEA.We have no known conflict of interest to disclose. This project was supported by an NSERC Discovery Grant (DGECR-2022-00309) and a Canada Research Chair (Tier II, CRC-2020-00174) to JAEA. CRediT Statement SA: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review and editing RP: Conceptualization, Investigation, Methodology, Software, Supervision, Validation, Writing – review and editing CM: Investigation, Methodology, Validation, Visualization, Writing - review and editing PZ: Investigation, Methodology, Validation, Visualization, Writing – review and editing JAE: Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – review and editing Acknowledgements Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly andCompany; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health ( www.fnih.org ). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California. Footnotes ↵ 5 Data used in preparation of this article were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database (adni.loni.usc.edu). As such, the investigators within the ADNI contributed to the design and implementation of ADNI and/or provided data but did not participate in analysis or writing of this report. A complete listing of ADNI investigators can be found at: http://adni.loni.usc.edu/wp-content/uploads/how_to_apply/ADNI_Acknowledgement_List.pdf ↵ 6 Among the analyzed models, 72% of results for the fixed effects of diagnosis, including the intercept, met significance criteria (p < 0.05) after applying the FDR correction. The significance ratios are as follows: Intercept (Stable CN) at 100%, Stable MCI at 78%, CN to MCI/AD at 78%, Stable AD 70%, and Progressive MCI to AD at 32%. The interactions of diagnosis and follow-up years yielded in total 67.6% and for the following terms: MCI to AD at 94%, CN to MCI/AD at 84%, Stable AD at 82%, Follow-Up Years at 70%, and Stable MCI at 8%. ↵ 7 Standardized residuals from the Chi-Square test showed significant gender distribution differences. In the Stable CN group, females had a residual of 41.99 (higher than expected), and males had −41.99 (lower than expected). In the Stable MCI group, females had −29.23 and males 29.23. The Stable AD group had minimal deviations (females: 5.26; males: −5.26). For the CN to MCI/AD transition, females had 9.09 and males −9.09. In the MCI to AD group, females had −25.33 and males 25.33. References ↵ Apostolova , L. G. , Dutton , R. A. , Dinov , I. D. , Hayashi , K. M. , Toga , A. W. , Cummings , J. L. , & Thompson , P. M . ( 2006 ). Conversion of mild cognitive impairment to Alzheimer disease predicted by hippocampal atrophy maps . Archives of Neurology , 63 ( 5 ), 693 – 699 . doi: 10.1001/archneur.63.5.693 OpenUrl CrossRef PubMed Web of Science ↵ Avants , B. B. , Epstein , C. L. , Grossman , M. , & Gee , J. C . ( 2008 ). Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain . Medical Image Analysis , 12 ( 1 ), 26 – 41 . doi: 10.1016/j.media.2007.06.004 OpenUrl CrossRef PubMed Web of Science ↵ Benjamini , Y. , & Hochberg , Y . ( 1995 ). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing . Journal of the Royal Statistical Society. Series B (Methodological) , 57 ( 1 ), 289 – 300 . OpenUrl CrossRef PubMed Web of Science ↵ Broadhouse , K. M. , Mowszowski , L. , Duffy , S. , Leung , I. , Cross , N. , Valenzuela , M. J. , & Naismith , S. L . ( 2019 ). Memory Performance Correlates of Hippocampal Subfield Volume in Mild Cognitive Impairment Subtype . Frontiers in Behavioral Neuroscience , 13 . doi: 10.3389/fnbeh.2019.00259 OpenUrl CrossRef PubMed ↵ Chang , C. , Huang , C. , Zhou , N. , Li , S. X. , Ver Hoef , L. , & Gao , Y . ( 2018 ). The bumps under the hippocampus . Human Brain Mapping , 39 ( 1 ), 472 – 490 . doi: 10.1002/hbm.23856 OpenUrl CrossRef PubMed ↵ Chow , N. , Hwang , K. S. , Hurtz , S. , Green , A. E. , Somme , J. H. , Thompson , P. M. , Elashoff , D. A. , Jack , C. R. , Weiner , M. , & Apostolova , L. G . ( 2015 ). Comparing 3T and 1.5T MRI for Mapping Hippocampal Atrophy in the Alzheimer’s Disease Neuroimaging Initiative . American Journal of Neuroradiology , 36 ( 4 ), 653 – 660 . doi: 10.3174/ajnr.A4228 OpenUrl Abstract / FREE Full Text ↵ Christopher-Hayes , N. J. , Embury , C. M. , Wiesman , A. I. , May , P. E. , Schantell , M. , Johnson , C. M. , Wolfson , S. L. , Murman , D. L. , & Wilson , T. W . ( 2023 ). Piecing it together: Atrophy profiles of hippocampal subfields relate to cognitive impairment along the Alzheimer’s disease spectrum . Frontiers in Aging Neuroscience , 15 . doi: 10.3389/fnagi.2023.1212197 OpenUrl CrossRef ↵ Coupé , P. , Eskildsen , S. F. , Manjón , J. V. , Fonov , V. S. , & Collins , D. L . ( 2012 ). Simultaneous segmentation and grading of anatomical structures for patient’s classification: Application to Alzheimer’s disease . NeuroImage , 59 ( 4 ), 3736 – 3747 . doi: 10.1016/j.neuroimage.2011.10.080 OpenUrl CrossRef PubMed ↵ Dale , A. M. , Fischl , B. , & Sereno , M. I . ( 1999 ). Cortical surface-based analysis: I. Segmentation and surface reconstruction . NeuroImage , 9 ( 2 ), 179 – 194 . OpenUrl CrossRef PubMed Web of Science ↵ DeKraker , J. , Lau , J. C. , Ferko , K. M. , Khan , A. R. , & Köhler , S . ( 2020 ). Hippocampal subfields revealed through unfolding and unsupervised clustering of laminar and morphological features in 3D BigBrain . NeuroImage , 206 , 116328 . doi: 10.1016/j.neuroimage.2019.116328 OpenUrl CrossRef ↵ DeKraker , J. , Köhler , S. , & Khan , A. R . ( 2021 ). Surface-based hippocampal subfield segmentation . Trends in Neurosciences , 44 ( 11 ), 856 – 863 . doi: 10.1016/j.tins.2021.06.005 OpenUrl CrossRef PubMed ↵ DeKraker , J. , Haast , R. A. M. , Yousif , M. D. , Karat , B. , Lau , J. C. , Köhler , S. , & Khan , A. R . ( 2022 ). Automated hippocampal unfolding for morphometry and subfield segmentation with HippUnfold . eLife , 11 , e77945 . doi: 10.7554/eLife.77945 OpenUrl CrossRef PubMed ↵ Duvernoy , H. M . ( 1998 ). The Human Hippocampus . Springer . doi: 10.1007/978-3-662-03628-0 OpenUrl CrossRef ↵ Eichenbaum , H . ( 2017 ). On the integration of space, time, and memory . Neuron 95 , 1007 – 1018 . doi: 10.1016/j.neuron.2017.06.036 OpenUrl CrossRef PubMed ↵ Faber , J. , Kügler , D. , Bahrami , E. , Heinz , L.-S. , Timmann , D. , Ernst , T. M. , Deike-Hofmann , K. , Klockgether , T. , van de Warrenburg , B. , van Gaalen , J. , Reetz , K. , Romanzetti , S. , Oz , G. , Joers , J. M. , Diedrichsen , J. , Giunti , P. , Garcia-Moreno , H. , Jacobi , H. , Jende , J. , … Reuter , M. ( 2022 ). CerebNet : A fast and reliable deep-learning pipeline for detailed cerebellum sub-segmentation . NeuroImage , 264 , 119703 . doi: 10.1016/j.neuroimage.2022.119703 OpenUrl CrossRef PubMed ↵ Fewtrell , M. S. , Kennedy , K. , Singhal , A. , Martin , R. M. , Ness , A. , Hadders-Algra , M. , et al. ( 2008 ). How much loss to follow-up is acceptable in long-term randomised trials and prospective studies? Archives of Disease in Childhood , 93 ( 6 ), 458 – 461 . OpenUrl FREE Full Text ↵ Fischl , B. , & Dale , A. M . ( 2000 ). Measuring the thickness of the human cerebral cortex from magnetic resonance images . Proceedings of the National Academy of Sciences , 97 ( 20 ), 11050 – 11055 . OpenUrl Abstract / FREE Full Text ↵ Goubran , M. , Ntiri , E. E. , Akhavein , H. , Holmes , M. , Nestor , S. , Ramirez , J. , Adamo , S. , Ozzoude , M. , Scott , C. , Gao , F. , Martel , A. , Swardfager , W. , Masellis , M. , Swartz , R. , MacIntosh , B. , & Black , S. E . ( 2020 ). Hippocampal segmentation for brains with extensive atrophy using three-dimensional convolutional neural networks . Human Brain Mapping , 41 ( 2 ), 291 – 308 . doi: 10.1002/hbm.24811 OpenUrl CrossRef PubMed ↵ Gupta , A. , Calfas , K. J. , Marshall , S. J. , Robinson , T. N. , Rock , C. L. , Huang , J. S. , et al. ( 2015 ). Clinical trial management of participant recruitment, enrollment, engagement, and retention in the SMART study using a Marketing and Information Technology (MARKIT) model . Contemporary Clinical Trials , 42 , 185 – 195 . OpenUrl PubMed ↵ Henschel , L. , Conjeti , S. , Estrada , S. , Diers , K. , Fischl , B. , & Reuter , M . ( 2020 ). Fastsurfer-a fast and accurate deep learning based neuroimaging pipeline . NeuroImage , 219 , 117012 . OpenUrl CrossRef PubMed ↵ Henschel , L. , Kügler , D. , & Reuter , M . ( 2022 ). FastSurferVINN: Building resolution-independence into deep learning segmentation methods-A solution for HighRes brain MRI . NeuroImage , 251 , 118933 . doi: 10.1016/j.neuroimage.2022.118933 OpenUrl CrossRef PubMed Hitti , F. L. , & Siegelbaum , S. A . ( 2014 ). The hippocampal CA2 region is essential for social memory . Nature , 508 ( 7494 ), 88 – 92 . OpenUrl CrossRef PubMed ↵ Ho , A. J. , Hua , X. , Lee , S. , Leow , A. D. , Yanovsky , I. , Gutman , B. , Dinov , I. D. , Leporé , N. , Stein , J. L. , Toga , A. W. , Jack Jr. , C. R. , Bernstein , M. A. , Reiman , E. M. , Harvey , D. J. , Kornak , J. , Schuff , N. , Alexander , G. E. , Werner , M. W. , & Thompson , P. M . ( 2010 ). Comparing 3 T and 1.5 T MRI for tracking Alzheimer’s disease progression with tensor-based morphometry . Human Brain Mapping , 31 ( 4 ), 499 – 514 . doi: 10.1002/hbm.20882 OpenUrl CrossRef PubMed ↵ Hojjati , S. H. , Butler , T. A. , Chiang , G. C. , Habeck , C. , RoyChoudhury , A. , Feiz , F. , Shteingart , J. , Nayak , S. , Ozoria , S. , Fernández , A. , Stern , Y. , Luchsinger , J. A. , Devanand , D. P. , & Razlighi , Q. R . ( 2023 ). Distinct and joint effects of low and high levels of Aβ and tau deposition on cortical thickness . NeuroImage : Clinical , 38 , 103409 . doi: 10.1016/j.nicl.2023.103409 OpenUrl CrossRef ↵ Insausti , R. , & Amaral , D. G . ( 2004 ). Hippocampal formation . The human nervous system , 2 , 871 – 914 . OpenUrl ↵ Insel , P. S. , Weiner , M. , Mackin , R. S. , Mormino , E. , Lim , Y. Y. , Stomrud , E. , Palmqvist , S. , Masters , C. L. , Maruff , P. T. , Hansson , O. , & Mattsson , N . ( 2019 ). Determining clinically meaningful decline in preclinical Alzheimer disease . Neurology , 93 ( 4 ), e322 – e333 . doi: 10.1212/WNL.0000000000007831 . PMID: 31289148 ; PMCID: PMC6669933 . OpenUrl CrossRef PubMed ↵ Jacobs , H. I. L. , Gronenschild , E. H. B. M. , Evers , E. A. T. , Ramakers , I. H. G. B. , Hofman , P. A. M. , Backes , W. H. , Jolles , J. , Verhey , F. R. J. , & Van Boxtel , M. P. J. ( 2015 ). Visuospatial processing in early Alzheimer’s disease: A multimodal neuroimaging study . Cortex , 64 , 394 – 406 . doi: 10.1016/j.cortex.2012.01.005 OpenUrl CrossRef PubMed ↵ La Joie , R. , Perrotin , A. , de La Sayette , V. , Egret , S. , Doeuvre , L. , Belliard , S. , Eustache , F. , Desgranges , B. , & Chételat , G. ( 2013 ). Hippocampal subfield volumetry in mild cognitive impairment, Alzheimer’s disease and semantic dementia . NeuroImage: Clinical , 3 , 155 – 162 . doi: 10.1016/j.nicl.2013.08.007 OpenUrl CrossRef PubMed ↵ Mak , E. , Su , L. , Williams , G. B. , Watson , R. , Firbank , M. , Blamire , A. , & O’Brien , J . ( 2016 ). Differential Atrophy of Hippocampal Subfields: A Comparative Study of Dementia with Lewy Bodies and Alzheimer Disease . The American Journal of Geriatric Psychiatry , 24 ( 2 ), 136 – 143 . doi: 10.1016/j.jagp.2015.06.006 OpenUrl CrossRef PubMed ↵ Modat , M. , Ridgway , G. R. , Taylor , Z. A. , Lehmann , M. , Barnes , J. , Hawkes , D. J. , Fox , N. C. , & Ourselin , S . ( 2010 ). Fast free-form deformation using graphics processing units . Computer Methods and Programs in Biomedicine , 98 ( 3 ), 278 – 284 . doi: 10.1016/j.cmpb.2009.09.002 OpenUrl CrossRef PubMed Web of Science ↵ Montal , V. , Vilaplana , E. , Alcolea , D. , Pegueroles , J. , Pasternak , O. , González-Ortiz , S. , Clarimón , J. , Carmona-Iragui , M. , Illán-Gala , I. , Morenas-Rodríguez , E. , Ribosa-Nogué , R. , Sala , I. , Sánchez-Saudinós , M.-B. , García-Sebastian , M. , Villanúa , J. , Izagirre , A. , Estanga , A. , Ecay-Torres , M. , Iriondo , A. , … Fortea , J . ( 2018 ). Cortical microstructural changes along the Alzheimer’s disease continuum . Alzheimer’s & Dementia , 14 ( 3 ), 340 – 351 . doi: 10.1016/j.jalz.2017.09.013 OpenUrl CrossRef PubMed ↵ Morrison , C. , Dadar , M. , Shafiee , N. , Collins , D. L. , & for Alzheimer’s Disease Neuroimaging Initiative . ( 2023a ). Hippocampal grading provides higher classification accuracy for those in the AD trajectory than hippocampal volume . Human Brain Mapping , 44 ( 12 ), 4623 – 4633 . doi: 10.1002/hbm.26407 OpenUrl CrossRef PubMed ↵ Morrison , C. , Dadar , M. , Shafiee , N. , Collins , D. L. , & For Alzheimer’s Disease Neuroimaging Initiative . ( 2023b ). The use of hippocampal grading as a biomarker for preclinical and prodromal Alzheimer’s disease . Human Brain Mapping , 44 ( 8 ), 3147 – 3157 . doi: 10.1002/hbm.26269 OpenUrl CrossRef PubMed ↵ Mukherjee , S. , Choi , S.-E. , Lee , M. L. , Scollard , P. , Trittschuh , E. H. , Mez , J. , Saykin , A. J. , Gibbons , L. E. , Sanders , R. E. , Zaman , A. F. , Teylan , M. A. , Kukull , W. A. , Barnes , L. L. , Bennett , D. A. , Lacroix , A. Z. , Larson , E. B. , Cuccaro , M. , Mercado , S. , Dumitrescu , L. , … Crane , P. K . ( 2023 ). Cognitive domain harmonization and cocalibration in studies of older adults . Neuropsychology , 37 ( 4 ), 409 – 423 . doi: 10.1037/neu0000835 OpenUrl CrossRef ↵ Núñez , C. , Callén , A. , Lombardini , F. , Compta , Y. , Stephan-Otto , C. , & Alzheimer’s Disease Neuroimaging Initiative . ( 2020 ). Different Cortical Gyrification Patterns in Alzheimer’s Disease and Impact on Memory Performance . Annals of Neurology , 88 ( 1 ), 67 – 80 . doi: 10.1002/ana.25741 OpenUrl CrossRef PubMed ↵ Olsen , R. K. , Carr , V. A. , Daugherty , A. M. , La Joie , R. , Amaral , R. S. , Amunts , K. , … & Hippocampal Subfields Group . ( 2019 ). Progress update from the hippocampal subfields group. Alzheimer’s & Dementia: Diagnosis , Assessment & Disease Monitoring , 11 ( 1 ), 439 – 449 . OpenUrl ↵ Pauli , W. M. , Nili , A. N. , & Tyszka , J. M . ( 2018 ). A high-resolution probabilistic in vivo atlas of human subcortical brain nuclei . Scientific Data , 5 ( 1 ), 180063 . doi: 10.1038/sdata.2018.63 OpenUrl CrossRef PubMed ↵ Phan , T.X. , Baratono , S. , Drew , W. , Tetreault , A.M. , Fox , M.D. , Darby , R.R. , and ( 2024 ), Increased Cortical Thickness in Alzheimer’s Disease . Ann Neurol . doi: 10.1002/ana.26894 OpenUrl CrossRef PubMed ↵ R Core Team ( 2021 ). R: A language and environment for statistical computing . R Foundation for Statistical Computing , Vienna, Austria . https://www.R-project.org/ . ↵ Rizzo , M. , Anderson , S. W. , Dawson , J. , & Nawrot , M . ( 2000 ). Vision and cognition in Alzheimer’s disease . Neuropsychologia , 38 ( 8 ), 1157 – 1169 . OpenUrl CrossRef PubMed Web of Science ↵ Schaer , M. , Cuadra , M. B. , Tamarit , L. , Lazeyras , F. , Eliez , S. , & Thiran , J. P . ( 2008 ). A surface-based approach to quantify local cortical gyrification . IEEE Transactions on Medical Imaging , 27 ( 2 ), 161 – 170 . OpenUrl CrossRef PubMed Web of Science ↵ Sghirripa , S. , Bhalerao , G. , Griffanti , L. , Gillis , G. , Mackay , C. , Voets , N. , Wong , S. , & Jenkinson , M. [Preprint]. ( 2024 ). Evaluating Traditional , Deep Learning, and Subfield Methods for Automatically Segmenting the Hippocampus from MRI. medRxiv , 2024.08.06.24311530. doi: 10.1101/2024.08.06.24311530 OpenUrl Abstract / FREE Full Text ↵ Sherif , T. , Rioux , P. , Rousseau , M.-E. , Kassis , N. , Beck , N. , Adalat , R. , Das , S. , Glatard , T. , & Evans , A. C . ( 2014 ). CBRAIN: A web-based, distributed computing platform for collaborative neuroimaging research . Frontiers in Neuroinformatics , 8 . doi: 10.3389/fninf.2014.00054 OpenUrl CrossRef PubMed ↵ Siddiqi , A. E. , Sikorskii , A. , Given , C. W. , & Given , B . ( 2008 ). Early participant attrition from clinical trials: role of trial design and logistics . Clinical Trials , 5 ( 4 ), 328 – 335 . OpenUrl PubMed ↵ Thyreau , B. , Sato , K. , Fukuda , H. , & Taki , Y . ( 2018 ). Segmentation of the hippocampus by transferring algorithmic knowledge for large cohort processing . Medical Image Analysis , 43 , 214 – 228 . doi: 10.1016/j.media.2017.11.004 OpenUrl CrossRef PubMed ↵ E. Tulving and W. Donaldson Tulving , E . ( 1972 ). “ Episodic and semantic memory ” in Organization of Memory . Eds. E. Tulving and W. Donaldson ( Cambridge, MA : Academic Press ). 381 – 403 . ↵ Van Essen , D. C. , & the Human Connectome Project . ( 2012 ). Connectome Workbench (Version 1.0) [Software] . Washington University in St. Louis ↵ Wisse , L. E. M. , Daugherty , A. M. , Olsen , R. K. , Berron , D. , Carr , V. A. , Stark , C. E. L. , Amaral , R. S. C. , Amunts , K. , Augustinack , J. C. , Bender , A. R. , Bernstein , J. D. , Boccardi , M. , Bocchetta , M. , Burggren , A. , Chakravarty , M. M. , Chupin , M. , Ekstrom , A. , de Flores , R. , Insausti , R. , … la Joie , R. ( 2017 ). A harmonized segmentation protocol for hippocampal and parahippocampal subregions: Why do we need one and what are the key goals? Hippocampus , 27 ( 1 ), 3 – 11 . doi: 10.1002/hipo.22671 OpenUrl CrossRef PubMed ↵ Yu , J . ( 2024 ). Age-related decline in thickness and surface area in the cortical surface and hippocampus: Lifespan trajectories and decade-by-decade analyses . GeroScience . doi: 10.1007/s11357-024-01220-1 OpenUrl CrossRef ↵ Yushkevich , P. A. , Piven , J. , Hazlett , H. C. , Smith , R. G. , Ho , S. , Gee , J. C. , & Gerig , G . ( 2006 ). User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability . NeuroImage , 31 ( 3 ), 1116 – 1128 . doi: 10.1016/j.neuroimage.2006.01.015 OpenUrl CrossRef PubMed Web of Science ↵ Yushkevich , P. A. , Amaral , R. S. , Augustinack , J. C. , Bender , A. R. , Bernstein , J. D. , Boccardi , M. , Bocchetta , M. , Burggren , A. C. , Carr , V. A. , Chakravarty , M. M. , Chételat , G. , Daugherty , A. M. , Davachi , L. , Ding , S. L. , Ekstrom , A. , Geerlings , M. I. , Hassan , A. , Huang , Y. , Iglesias , J. E. , La Joie , R ., … Hippocampal Subfields Group (HSG) ( 2015 ). Quantitative comparison of 21 protocols for labeling hippocampal subfields and parahippocampal subregions in in vivo MRI: towards a harmonized segmentation protocol . NeuroImage , 111 , 526 – 541 . doi: 10.1016/j.neuroimage.2015.01.004 OpenUrl CrossRef PubMed ↵ Zhao W. , Wang X. , Yin C. , He M. , Li S. , Han Y . ( 2019 . ) Trajectories of the Hippocampal Subfields Atrophy in the Alzheimer’s Disease: A Structural Imaging Study . Frontiers in Neuroinformatics ; 13 . View the discussion thread. Back to top Previous Next Posted April 08, 2025. Download PDF Supplementary Material Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Longitudinal Surface-Based Morphometry Changes in the Hippocampus in Dementia Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Longitudinal Surface-Based Morphometry Changes in the Hippocampus in Dementia Salah Aziz , Romeo Penheiro , Cassandra Morrison , Peter Zhukovsky , John AE Anderson , the Alzheimer’s Disease Neuroimaging Initiative (ADNI) medRxiv 2025.04.07.25325336; doi: https://doi.org/10.1101/2025.04.07.25325336 Share This Article: Copy Citation Tools Longitudinal Surface-Based Morphometry Changes in the Hippocampus in Dementia Salah Aziz , Romeo Penheiro , Cassandra Morrison , Peter Zhukovsky , John AE Anderson , the Alzheimer’s Disease Neuroimaging Initiative (ADNI) medRxiv 2025.04.07.25325336; doi: https://doi.org/10.1101/2025.04.07.25325336 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Radiology and Imaging Subject Areas All Articles Addiction Medicine (568) Allergy and Immunology (863) Anesthesia (300) Cardiovascular Medicine (4435) Dentistry and Oral Medicine (444) Dermatology (382) Emergency Medicine (608) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1509) Epidemiology (15228) Forensic Medicine (30) Gastroenterology (1124) Genetic and Genomic Medicine (6598) Geriatric Medicine (668) Health Economics (997) Health Informatics (4536) Health Policy (1368) Health Systems and Quality Improvement (1613) Hematology (540) HIV/AIDS (1264) Infectious Diseases (except HIV/AIDS) (15916) Intensive Care and Critical Care Medicine (1103) Medical Education (623) Medical Ethics (146) Nephrology (667) Neurology (6599) Nursing (346) Nutrition (998) Obstetrics and Gynecology (1144) Occupational and Environmental Health (957) Oncology (3332) Ophthalmology (974) Orthopedics (369) Otolaryngology (420) Pain Medicine (436) Palliative Medicine (130) Pathology (663) Pediatrics (1693) Pharmacology and Therapeutics (691) Primary Care Research (711) Psychiatry and Clinical Psychology (5447) Public and Global Health (9230) Radiology and Imaging (2198) Rehabilitation Medicine and Physical Therapy (1370) Respiratory Medicine (1196) Rheumatology (593) Sexual and Reproductive Health (712) Sports Medicine (530) Surgery (712) Toxicology (99) Transplantation (289) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a0051e26a80f41e2',t:'MTc3OTU1MDM5Mw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();