Controlled environment agriculture is a promising solution to address climate change and resource limitations. Light, the primary energy source driving photosynthesis and regulating plant growth, is critical in optimizing produce quality. However, the impact of specific light spectra during night interruption on improving phytochemical content and produce quality remains underexplored. This study investigated the effects of red (peak wavelength at 660 nm) and far-red night interruption (peak wavelength at 730 nm) on photosynthetic efficiency, biomass distribution, and phytochemical production in Italian basil (Ocimum basilicum L.). Treatments included red light, far-red light, a combination of both, and a control without night interruption. Red light significantly increased chlorophyll a by 16.8%, chlorophyll b by 20.6%, and carotenoids by 11%, improving photosynthetic efficiency and nutritional quality. Red light also elevated anthocyanin levels by 15.5%, while far-red light promoted flavonoid production by 43.56%. Although red light enhanced biomass, the primary benefit was improved leaf quality, with more biomass directed to leaves over roots. Far-red light reduced transpiration, enhancing post-harvest water retention and shelf life. These findings demonstrate that red and far-red night interruption can optimize phytochemical content, produce quality, and post-harvest durability, offering valuable insights for controlled environment agriculture. Future research should focus on refining night interruption light strategies across a broader range of crops to enhance produce quality and shelf life in controlled environment agriculture.

Alzheimer’s disease (AD) detection from structural magnetic resonance imaging (MRI) using deep learning depends on large, labelled datasets. However, many cohorts contain only a few hundred subjects, for which standard augmentation adds very limited diversity. Latent diffusion models (LDMs) offer an alternative. The current study presents a reproducible two-stage pipeline that generates 2D coronal brain MRI conditioned on diagnosis and trained on 295 subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). In the first stage, a variational autoencoder with an adversarial objective compressed 256 × 256 slices into a 32 × 32 × 8 latent space. Then, a class-conditional LDM with classifier-free guidance generated AD and cognitively normal (CN) images. A baseline and a redesigned pipeline adding MNI152 registration, anatomically guided slice selection, and increased latent capacity were compared on the same subjects. The redesigned pipeline reached a Kernel Inception Distance of 0.030 ± 0.002 and a bias-corrected Fréchet Inception Distance (FID∞) of 42.14. A classifier trained only on synthetic data achieved an area under the curve (AUC) of 0.754 (95% CI, 0.60 to 0.91), compared with 0.810 for real data. No instance memorisation was detected across 880 samples, and a control analysis exposed a 32.7-percentage-point inflation in the standard memorisation metric under unequal reference sets. The redesigned preprocessing pipeline improves synthesis quality at a small-cohort scale and offers an approach that may be transferable to other privacy-restricted medical imaging domains.

Deep learning for Alzheimer’s disease (AD) detection from structural magnetic resonance imaging (MRI) needs large, labelled datasets, yet many cohorts hold only a few hundred participants, for which conventional augmentation adds little anatomical diversity. In a two-stage pipeline, a variational autoencoder compressed 256 × 256 coronal slices to a 32 × 32 × 8 latent space, and a class-conditional latent diffusion model under classifier-free guidance generated AD and cognitively normal (CN) images using 295 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). The pipeline reached a Kernel Inception Distance (KID) of 0.030 ± 0.002 and a bias-corrected Fréchet Inception Distance (FID_∞) of 43.82. A controlled ablation varying preprocessing alone improved KID by 0.0147 and precision by 0.069, both with 95% intervals excluding zero. FID did not separate the configurations. A ResNet-18 trained only on synthetic slices and tested on 44 held-out real participants (18 AD, 26 CN), each scored as the mean probability over twenty slices, reached an area under the curve of 0.779 ± 0.031 against 0.869 ± 0.027 for real data; the difference was not distinguishable at this sample size. No instance memorisation was found among 880 samples, and a size-matched control exposed a 27.7-percentage-point inflation in the standard memorisation metric. Preprocessing, therefore, measurably affects synthesis quality at the small-cohort scale, though not on every measure.