Distinct spatiotemporal atrophy patterns in corticobasal syndrome are associated with different underlying pathologies

Objective

To identify imaging subtypes of the cortico-basal syndrome (CBS) based solely on a data-driven assessment of MRI atrophy patterns, and investigate whether these subtypes provide information on the underlying pathology.

Methods

We applied Subtype and Stage Inference (SuStaIn), a machine learning algorithm that identifies groups of individuals with distinct biomarker progression patterns, to a large cohort of 135 CBS cases (52 had a pathological or biomarker defined diagnosis) and 252 controls. The model was fit using volumetric features extracted from baseline T1-weighted MRI scans and validated using follow-up MRI. We compared the clinical phenotypes of each subtype and investigated whether there were differences in associated pathology between the subtypes.

Results

SuStaIn identified two subtypes with distinct sequences of atrophy progression; four-repeat-tauopathy confirmed cases were most commonly assigned to the Subcortical subtype (83% of CBS-PSP and 75% of CBS-CBD), while CBS-AD was most commonly assigned to the Fronto-parieto-occipital subtype (81% of CBS-AD). Subtype assignment was stable at follow-up (98% of cases), and individuals consistently progressed to higher stages (100% stayed at the same stage or progressed), supporting the model's ability to stage progression.

Interpretation

By jointly modelling disease stage and subtype, we provide data-driven evidence for at least two distinct and longitudinally stable spatiotemporal subtypes of atrophy in CBS that are associated with different underlying pathologies. In the absence of sensitive and specific biomarkers, accurately subtyping and staging individuals with CBS at baseline has important implications for screening on entry into clinical trials, as well as for tracking disease progression.

Artificial intelligence for dementia research methods optimization

Artificial intelligence (AI) and machine learning (ML) approaches are increasingly being used in dementia research. However, several methodological challenges exist that may limit the insights we can obtain from high-dimensional data and our ability to translate these findings into improved patient outcomes. To improve reproducibility and replicability, researchers should make their well-documented code and modeling pipelines openly available. Data should also be shared where appropriate. To enhance the acceptability of models and AI-enabled systems to users, researchers should prioritize interpretable methods that provide insights into how decisions are generated. Models should be developed using multiple, diverse datasets to improve robustness, generalizability, and reduce potentially harmful bias. To improve clarity and reproducibility, researchers should adhere to reporting guidelines that are co-produced with multiple stakeholders. If these methodological challenges are overcome, AI and ML hold enormous promise for changing the landscape of dementia research and care. HIGHLIGHTS: Machine learning (ML) can improve diagnosis, prevention, and management of dementia. Inadequate reporting of ML procedures affects reproduction/replication of results. ML models built on unrepresentative datasets do not generalize to new datasets. Obligatory metrics for certain model structures and use cases have not been defined. Interpretability and trust in ML predictions are barriers to clinical translation.

Artificial Intelligence for Dementia Research Methods Optimization

INTRODUCTION

Machine learning (ML) has been extremely successful in identifying key features from high-dimensional datasets and executing complicated tasks with human expert levels of accuracy or greater.

METHODS

We summarize and critically evaluate current applications of ML in dementia research and highlight directions for future research.

RESULTS

We present an overview of ML algorithms most frequently used in dementia research and highlight future opportunities for the use of ML in clinical practice, experimental medicine, and clinical trials. We discuss issues of reproducibility, replicability and interpretability and how these impact the clinical applicability of dementia research. Finally, we give examples of how state-of-the-art methods, such as transfer learning, multi-task learning, and reinforcement learning, may be applied to overcome these issues and aid the translation of research to clinical practice in the future.

DISCUSSION

ML-based models hold great promise to advance our understanding of the underlying causes and pathological mechanisms of dementia.

Uncovering spatiotemporal patterns of atrophy in progressive supranuclear palsy using unsupervised machine learning

To better understand the pathological and phenotypic heterogeneity of progressive supranuclear palsy and the links between the two, we applied a novel unsupervised machine learning algorithm (Subtype and Stage Inference) to the largest MRI data set to date of people with clinically diagnosed progressive supranuclear palsy (including progressive supranuclear palsy-Richardson and variant progressive supranuclear palsy syndromes). Our cohort is comprised of 426 progressive supranuclear palsy cases, of which 367 had at least one follow-up scan, and 290 controls. Of the progressive supranuclear palsy cases, 357 were clinically diagnosed with progressive supranuclear palsy-Richardson, 52 with a progressive supranuclear palsy-cortical variant (progressive supranuclear palsy-frontal, progressive supranuclear palsy-speech/language, or progressive supranuclear palsy-corticobasal), and 17 with a progressive supranuclear palsy-subcortical variant (progressive supranuclear palsy-parkinsonism or progressive supranuclear palsy-progressive gait freezing). Subtype and Stage Inference was applied to volumetric MRI features extracted from baseline structural (T1-weighted) MRI scans and then used to subtype and stage follow-up scans. The subtypes and stages at follow-up were used to validate the longitudinal consistency of subtype and stage assignments. We further compared the clinical phenotypes of each subtype to gain insight into the relationship between progressive supranuclear palsy pathology, atrophy patterns, and clinical presentation. The data supported two subtypes, each with a distinct progression of atrophy: a 'subcortical' subtype, in which early atrophy was most prominent in the brainstem, ventral diencephalon, superior cerebellar peduncles, and the dentate nucleus, and a 'cortical' subtype, in which there was early atrophy in the frontal lobes and the insula alongside brainstem atrophy. There was a strong association between clinical diagnosis and the Subtype and Stage Inference subtype with 82% of progressive supranuclear palsy-subcortical cases and 81% of progressive supranuclear palsy-Richardson cases assigned to the subcortical subtype and 82% of progressive supranuclear palsy-cortical cases assigned to the cortical subtype. The increasing stage was associated with worsening clinical scores, whilst the 'subcortical' subtype was associated with worse clinical severity scores compared to the 'cortical subtype' (progressive supranuclear palsy rating scale and Unified Parkinson's Disease Rating Scale). Validation experiments showed that subtype assignment was longitudinally stable (95% of scans were assigned to the same subtype at follow-up) and individual staging was longitudinally consistent with 90% remaining at the same stage or progressing to a later stage at follow-up. In summary, we applied Subtype and Stage Inference to structural MRI data and empirically identified two distinct subtypes of spatiotemporal atrophy in progressive supranuclear palsy. These image-based subtypes were differentially enriched for progressive supranuclear palsy clinical syndromes and showed different clinical characteristics. Being able to accurately subtype and stage progressive supranuclear palsy patients at baseline has important implications for screening patients on entry to clinical trials, as well as tracking disease progression.

Heterogeneity in Preclinical Alzheimer's Disease Trial Cohort Identified by Image-based Data-Driven Disease Progression Modelling

Importance

Undetected biological heterogeneity adversely impacts trials in Alzheimer's disease because rate of cognitive decline - and perhaps response to treatment - differs in subgroups. Recent results show that data-driven approaches can unravel the heterogeneity of Alzheimer's disease progression. The resulting stratification is yet to be leveraged in clinical trials.

Objective

Investigate whether image-based data-driven disease progression modelling could identify baseline biological heterogeneity in a clinical trial, and whether these subgroups have prognostic or predictive value.

Design

Screening data from the Anti-Amyloid Treatment in Asymptomatic Alzheimer Disease (A4) Study collected between April 2014 and December 2017, and longitudinal data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) observational study downloaded in February 2022 were used.

Setting

The A4 Study is an interventional trial involving 67 sites in the US, Canada, Australia, and Japan. ADNI is a multi-center observational study in North America.

Participants

Cognitively unimpaired amyloid-positive participants with a 3-Tesla T1-weighted MRI scan. Amyloid positivity was determined using florbetapir PET imaging (in A4) and CSF Aβ(1-42) (in ADNI).

Main Outcomes and Measures

Regional volumes estimated from MRI scans were used as input to the Subtype and Stage Inference (SuStaIn) algorithm. Outcomes included cognitive test scores and SUVr values from florbetapir and flortaucipir PET.

Results

We included 1,240 Aβ+ participants (and 407 Aβ- controls) from the A4 Study, and 731 A4-eligible ADNI participants. SuStaIn identified three neurodegeneration subtypes - Typical, Cortical, Subcortical - comprising 523 (42%) individuals. The remainder are designated subtype zero (insufficient atrophy). Baseline PACC scores (A4 primary outcome) were significantly worse in the Cortical subtype (median = -1.27, IQR=[-3.34,0.83]) relative to both subtype zero (median=-0.013, IQR=[-1.85,1.67], P<.0001) and the Subcortical subtype (median=0.03, IQR=[-1.78,1.61], P=.0006). In ADNI, over a four-year period (comparable to A4), greater cognitive decline in the mPACC was observed in both the Typical (-0.23/yr; 95% CI, [-0.41,-0.05]; P=.01) and Cortical (-0.24/yr; [-0.42,-0.06]; P=.009) subtypes, as well as the CDR-SB (Typical: +0.09/yr, [0.06,0.12], P<.0001; and Cortical: +0.07/yr, [0.04,0.10], P<.0001).

Conclusions and Relevance

In a large secondary prevention trial, our image-based model detected neurodegenerative heterogeneity predictive of cognitive heterogeneity. We argue that such a model is a valuable tool to be considered in future trial design to control for previously undetected variance.

Targeted Screening for Alzheimer's Disease Clinical Trials Using Data-Driven Disease Progression Models

Heterogeneity in Alzheimer's disease progression contributes to the ongoing failure to demonstrate efficacy of putative disease-modifying therapeutics that have been trialed over the past two decades. Any treatment effect present in a subgroup of trial participants (responders) can be diluted by non-responders who ideally should have been screened out of the trial. How to identify (screen-in) the most likely potential responders is an important question that is still without an answer. Here, we pilot a computational screening tool that leverages recent advances in data-driven disease progression modeling to improve stratification. This aims to increase the sensitivity to treatment effect by screening out non-responders, which will ultimately reduce the size, duration, and cost of a clinical trial. We demonstrate the concept of such a computational screening tool by retrospectively analyzing a completed double-blind clinical trial of donepezil in people with amnestic mild cognitive impairment (clinicaltrials.gov: NCT00000173), identifying a data-driven subgroup having more severe cognitive impairment who showed clearer treatment response than observed for the full cohort.

pySuStaIn: a Python implementation of the Subtype and Stage Inference algorithm

Progressive disorders are highly heterogeneous. Symptom-based clinical classification of these disorders may not reflect the underlying pathobiology. Data-driven subtyping and staging of patients has the potential to disentangle the complex spatiotemporal patterns of disease progression. Tools that enable this are in high demand from clinical and treatment-development communities. Here we describe the pySuStaIn software package, a Python-based implementation of the Subtype and Stage Inference (SuStaIn) algorithm. SuStaIn unravels the complexity of heterogeneous diseases by inferring multiple disease progression patterns (subtypes) and individual severity (stages) from cross-sectional data. The primary aims of pySuStaIn are to enable widespread application and translation of SuStaIn via an accessible Python package that supports simple extension and generalization to novel modelling situations within a single, consistent architecture.

Explainable AI reveals changes in skin microbiome composition linked to phenotypic differences

Alterations in the human microbiome have been observed in a variety of conditions such as asthma, gingivitis, dermatitis and cancer, and much remains to be learned about the links between the microbiome and human health. The fusion of artificial intelligence with rich microbiome datasets can offer an improved understanding of the microbiome’s role in human health. To gain actionable insights it is essential to consider both the predictive power and the transparency of the models by providing explanations for the predictions. We combine the collection of leg skin microbiome samples from two healthy cohorts of women with the application of an explainable artificial intelligence (EAI) approach that provides accurate predictions of phenotypes with explanations. The explanations are expressed in terms of variations in the relative abundance of key microbes that drive the predictions. We predict skin hydration, subject's age, pre/post-menopausal status and smoking status from the leg skin microbiome. The changes in microbial composition linked to skin hydration can accelerate the development of personalized treatments for healthy skin, while those associated with age may offer insights into the skin aging process. The leg microbiome signatures associated with smoking and menopausal status are consistent with previous findings from oral/respiratory tract microbiomes and vaginal/gut microbiomes respectively. This suggests that easily accessible microbiome samples could be used to investigate health-related phenotypes, offering potential for non-invasive diagnosis and condition monitoring. Our EAI approach sets the stage for new work focused on understanding the complex relationships between microbial communities and phenotypes. Our approach can be applied to predict any condition from microbiome samples and has the potential to accelerate the development of microbiome-based personalized therapeutics and non-invasive diagnostics.

Sequential Nature of (p,3p) Two-Proton Knockout from Neutron-Rich Nuclei

Twenty-one two-proton knockout (p,3p) cross sections were measured from neutron-rich nuclei at ∼250  MeV/nucleon in inverse kinematics. The angular distribution of the three emitted protons was determined for the first time, demonstrating that the (p,3p) kinematics are consistent with two sequential proton-proton collisions within the projectile nucleus. Ratios of (p,3p) over (p,2p) inclusive cross sections follow the trend of other many-nucleon removal reactions, further reinforcing the sequential nature of (p,3p) in neutron-rich nuclei.

Metastable States of ^{92,94}Se: Identification of an Oblate K Isomer of ^{94}Se and the Ground-State Shape Transition between N=58 and 60

Here we present new information on the shape evolution of the very neutron-rich ^{92,94}Se nuclei from an isomer-decay spectroscopy experiment at the Radioactive Isotope Beam Factory at RIKEN. High-resolution germanium detectors were used to identify delayed γ rays emitted following the decay of their isomers. New transitions are reported extending the previously known level schemes. The isomeric levels are interpreted as originating from high-K quasineutron states with an oblate deformation of β∼0.25, with the high-K state in ^{94}Se being metastable and K hindered. Following this, ^{94}Se is the lowest-mass neutron-rich nucleus known to date with such a substantial K hindrance. Furthermore, it is the first observation of an oblate K isomer in a deformed nucleus. This opens up the possibility for a new region of K isomers at low Z and at oblate deformation, involving the same neutron orbitals as the prolate orbitals within the classic Z∼72 deformed hafnium region. From an interpretation of the level scheme guided by theoretical calculations, an oblate deformation is also suggested for the ^{94}Se_{60} ground-state band.

Prominence of Pairing in Inclusive (p,2p) and (p,pn) Cross Sections from Neutron-Rich Nuclei

Fifty-five inclusive single nucleon-removal cross sections from medium mass neutron-rich nuclei impinging on a hydrogen target at ∼250  MeV/nucleon are measured at the RIKEN Radioactive Isotope Beam Factory. Systematically higher cross sections are found for proton removal from nuclei with an even number of protons as compared to odd-proton number projectiles for a given neutron separation energy. Neutron removal cross sections display no even-odd splitting, contrary to nuclear cascade model predictions. Both effects are understood through simple considerations of neutron separation energies and bound state level densities originating in pairing correlations in the daughter nuclei. These conclusions are supported by comparison with semimicroscopic model predictions, highlighting the enhanced role of low-lying level densities in nucleon-removal cross sections from loosely bound nuclei.

Towards Neutron Capture on Exotic Nuclei: Demonstrating (d,pγ) as a Surrogate Reaction for (n,γ)

The neutron-capture reaction plays a critical role in the synthesis of the elements in stars and is important for societal applications including nuclear power generation and stockpile-stewardship science. However, it is difficult-if not impossible-to directly measure neutron capture cross sections for the exotic, short-lived nuclei that participate in these processes. In this Letter we demonstrate a new technique which can be used to indirectly determine neutron-capture cross sections for exotic systems. This technique makes use of the (d,p) transfer reaction, which has long been used as a tool to study the structure of nuclei. Recent advances in reaction theory, together with data collected using this reaction, enable the determination of neutron-capture cross sections for short-lived nuclei. A benchmark study of the ^{95}Mo(d,p) reaction is presented, which illustrates the approach and provides guidance for future applications of the method with short-lived isotopes produced at rare isotope accelerators.

Erratum: Persistence of the Z=28 Shell Gap Around ^{78}Ni: First Spectroscopy of ^{79}Cu [Phys. Rev. Lett. 119, 192501 (2017)]

This corrects the article DOI: 10.1103/PhysRevLett.119.192501.

Persistence of the Z=28 Shell Gap Around ^{78}Ni: First Spectroscopy of ^{79}Cu

In-beam γ-ray spectroscopy of ^{79}Cu is performed at the Radioactive Isotope Beam Factory of RIKEN. The nucleus of interest is produced through proton knockout from a ^{80}Zn beam at 270  MeV/nucleon. The level scheme up to 4.6 MeV is established for the first time and the results are compared to Monte Carlo shell-model calculations. We do not observe significant knockout feeding to the excited states below 2.2 MeV, which indicates that the Z=28 gap at N=50 remains large. The results show that the ^{79}Cu nucleus can be described in terms of a valence proton outside a ^{78}Ni core, implying the magic character of the latter.

Shape Evolution in Neutron-Rich Krypton Isotopes Beyond N=60: First Spectroscopy of ^{98,100}Kr

We report on the first γ-ray spectroscopy of low-lying states in neutron-rich ^{98,100}Kr isotopes obtained from ^{99,101}Rb(p,2p) reactions at ∼220  MeV/nucleon. A reduction of the 2_{1}^{+} state energies beyond N=60 demonstrates a significant increase of deformation, shifted in neutron number compared to the sharper transition observed in strontium and zirconium isotopes. State-of-the-art beyond-mean-field calculations using the Gogny D1S interaction predict level energies in good agreement with experimental results. The identification of a low-lying (0_{2}^{+}, 2_{2}^{+}) state in ^{98}Kr provides the first experimental evidence of a competing configuration at low energy in neutron-rich krypton isotopes consistent with the oblate-prolate shape coexistence picture suggested by theory.

94 β-Decay Half-Lives of Neutron-Rich _{55}Cs to _{67}Ho: Experimental Feedback and Evaluation of the r-Process Rare-Earth Peak Formation

The β-decay half-lives of 94 neutron-rich nuclei ^{144-151}Cs, ^{146-154}Ba, ^{148-156}La, ^{150-158}Ce, ^{153-160}Pr, ^{156-162}Nd, ^{159-163}Pm, ^{160-166}Sm, ^{161-168}Eu, ^{165-170}Gd, ^{166-172}Tb, ^{169-173}Dy, ^{172-175}Ho, and two isomeric states ^{174m}Er, ^{172m}Dy were measured at the Radioactive Isotope Beam Factory, providing a new experimental basis to test theoretical models. Strikingly large drops of β-decay half-lives are observed at neutron-number N=97 for _{58}Ce, _{59}Pr, _{60}Nd, and _{62}Sm, and N=105 for _{63}Eu, _{64}Gd, _{65}Tb, and _{66}Dy. Features in the data mirror the interplay between pairing effects and microscopic structure. r-process network calculations performed for a range of mass models and astrophysical conditions show that the 57 half-lives measured for the first time play an important role in shaping the abundance pattern of rare-earth elements in the solar system.

Are There Signatures of Harmonic Oscillator Shells Far from Stability? First Spectroscopy of ^{110}Zr

The first measurement of the low-lying states of the neutron-rich ^{110}Zr and ^{112}Mo was performed via in-beam γ-ray spectroscopy after one proton removal on hydrogen at ∼200  MeV/nucleon. The 2_{1}^{+} excitation energies were found at 185(11) keV in ^{110}Zr, and 235(7) keV in ^{112}Mo, while the R_{42}=E(4_{1}^{+})/E(2_{1}^{+}) ratios are 3.1(2), close to the rigid rotor value, and 2.7(1), respectively. These results are compared to modern energy density functional based configuration mixing models using Gogny and Skyrme effective interactions. We conclude that first levels of ^{110}Zr exhibit a rotational behavior, in agreement with previous observations of lighter zirconium isotopes as well as with the most advanced Monte Carlo shell model predictions. The data, therefore, do not support a harmonic oscillator shell stabilization scenario at Z=40 and N=70. The present data also invalidate predictions for a tetrahedral ground state symmetry in ^{110}Zr.

Extension of the N=40 Island of Inversion towards N=50: Spectroscopy of (66)Cr, (70,72)Fe

We report on the measurement of the first 2(+) and 4(+) states of (66)Cr and (70,72)Fe via in-beam γ-ray spectroscopy. The nuclei of interest were produced by (p,2p) reactions at incident energies of 260  MeV/nucleon. The experiment was performed at the Radioactive Isotope Beam Factory, RIKEN, using the DALI 2γ-ray detector array and the novel MINOS device, a thick liquid hydrogen target combined with a vertex tracker. A low-energy plateau of 2(1)(+) and 4(1)(+) energies as a function of the neutron number was observed for N≥38 and N≥40 for even-even Cr and Fe isotopes, respectively. State-of-the-art shell model calculations with a modified Lenzi-Nowacki-Poves-Sieja (LNPS) interaction in the pfg(9/2)d(5/2) valence space reproduce the observations. Interpretation within the shell model shows an extension of the island of inversion at N=40 for more neutron-rich isotopes towards N=50.

Soil- and plant-based countermeasures to reduce 137Cs and 90Sr uptake by grasses in natural meadows: the REDUP project

The effectiveness of a set of soil- and plant-based countermeasures to reduce 137Cs and 90Sr transfer to plants was tested in natural meadows in the area affected by Chernobyl fallout. Countermeasures comprised the use of agricultural practices (disking + ploughing, liming and NPK fertilisation), addition of soil amendments and reseeding with a selection of grass species. Disking + ploughing was the most effective treatment, whereas the K fertiliser doses applied were insufficient to produce a significant increase in K concentration in soil solution. The application of some agricultural practices was economically justifiable for scenarios with a high initial transfer, such as 137Cs-contaminated organic soils. The use of soil amendments did not lead to a further decrease in transfer. Laboratory experiments demonstrated that this was because of their low radionuclide sorption properties. Finally, experiments examining the effect of plant species on radionuclide transfer showed that both transfer and biomass can depend on the plant species, indicating that those with high radionuclide root uptake should be avoided when reseeding after ploughing.