Open Access

Isolated interacting quantum systems generally thermalize, yet there are several examples for the breakdown of ergodicity, such as many-body localization and quantum scars. Recently, ergodicity breaking has been observed in systems subjected to linear potentials, termed Stark many-body localization. This phenomenon is closely associated with Hilbert-space fragmentation, characterized by a strong dependence of dynamics on initial conditions. Here, we explore initial-state-dependent dynamics using a ladder-type superconducting processor with up to 24 qubits, which enables precise control of the qubit frequency and initial-state preparation. In systems with linear potentials, we experimentally observe distinct nonequilibrium dynamics for initial states with the same quantum numbers and energy, but with varying domain-wall numbers. Accompanied by the numerical simulation for systems with larger sizes, we reveal that this distinction becomes increasingly pronounced as the system size grows, in contrast with weakly disordered interacting systems. Our results provide convincing experimental evidence of the fragmentation in Stark systems, enriching our understanding of the weak breakdown of ergodicity.

Background Ga-[1,4,7,10-tetraazacyclododecane-N,N′,N″,N′″- tetraacetic acid]-d-Phe1,Tyr3-octreotate (DOTATATE) positron emission tomography (PET) is commonly used for the visualization of somatostatin receptor (SSTR)-positive neuroendocrine tumors. SSTR is also known to be expressed on macrophages, which play a major role in inflammatory processes in the walls of coronary arteries and large vessels. Therefore, imaging SSTR expression has the potential to visualize vulnerable plaques. We assessed 68Ga-DOTATATE accumulation in large vessels in comparison to 18F-2-fluorodeoxyglucose (FDG) uptake, calcified plaques (CPs), and cardiovascular risk factors. Methods Sixteen consecutive patients with neuroendocrine tumors or thyroid cancer underwent both 68Ga-DOTATATE and 18F-FDG PET/CT for staging or restaging purposes. Detailed clinical data, including common cardiovascular risk factors, were recorded. For a separate assessment, they were divided into a high-risk and a low-risk group. In each patient, we calculated the maximum target-to-background ratio (TBR) of eight arterial segments. The correlation of the TBRmean of both tracers with risk factors including plaque burden was assessed. Results The mean TBR of 68Ga-DOTATATE in all large arteries correlated significantly with the presence of CPs (r = 0.52; p < 0.05), hypertension (r = 0.60; p < 0.05), age (r = 0.56; p < 0.05), and uptake of 18F-FDG (r = 0.64; p < 0.01). There was one significant correlation between 18F-FDG uptake and hypertension (0.58; p < 0.05). Out of the 37 sites with the highest focal 68Ga-DOTATATE uptake, 16 (43.2%) also had focal 18F-FDG uptake. Of 39 sites with the highest 18F-FDG uptake, only 11 (28.2%) had a colocalized 68Ga-DOTATATE accumulation. Conclusions In this series of cancer patients, we found a stronger association of increased 68Ga-DOTATATE uptake with known risk factors of cardiovascular disease as compared to 18F-FDG, suggesting a potential role for plaque imaging in large arteries. Strikingly, we found that focal uptake of 68Ga-DOTATATE and 18F-FDG does not colocalize in a significant number of lesions.

Accurate risk prediction for myocardial infarction (MI) is crucial for preventive strategies, given its significant impact on global mortality and morbidity. Here, we propose a novel deep-learning approach to enhance the prediction of incident MI cases by incorporating metabolomics alongside clinical risk factors. We utilized data from the KORA cohort, including the baseline S4 and follow-up F4 studies, consisting of 1454 participants without prior history of MI. The dataset comprised 19 clinical variables and 363 metabolites. Due to the imbalanced nature of the dataset (78 observed MI cases and 1376 non-MI individuals), we employed a generative adversarial network (GAN) model to generate new incident cases, augmenting the dataset and improving feature representation. To predict MI, we further utilized multi-layer perceptron (MLP) models in conjunction with the synthetic minority oversampling technique (SMOTE) and edited nearest neighbor (ENN) methods to address overfitting and underfitting issues, particularly when dealing with imbalanced datasets. To enhance prediction accuracy, we propose a novel GAN for feature-enhanced (GFE) loss function. The GFE loss function resulted in an approximate 2% improvement in prediction accuracy, yielding a final accuracy of 70%. Furthermore, we evaluated the contribution of each clinical variable and metabolite to the predictive model and identified the 10 most significant variables, including glucose tolerance, sex, and physical activity. This is the first study to construct a deep-learning approach for producing 7-year MI predictions using the newly proposed loss function. Our findings demonstrate the promising potential of our technique in identifying novel biomarkers for MI prediction.