Neuron-level explainable AI for Alzheimer's Disease assessment from fundus images

Alzheimer's Disease (AD) is a progressive neurodegenerative disease and the leading cause of dementia. Early diagnosis is critical for patients to benefit from potential intervention and treatment. The retina has emerged as a plausible diagnostic site for AD detection owing to its anatomical connection with the brain. However, existing AI models for this purpose have yet to provide a rational explanation behind their decisions and have not been able to infer the stage of the disease's progression. Along this direction, we propose a novel model-agnostic explainable-AI framework, called Granula ̲ r Neuron-le v ̲ el Expl a ̲ iner (LAVA), an interpretation prototype that probes into intermediate layers of the Convolutional Neural Network (CNN) models to directly assess the continuum of AD from the retinal imaging without the need for longitudinal or clinical evaluations. This innovative approach aims to validate retinal vasculature as a biomarker and diagnostic modality for evaluating Alzheimer's Disease. Leveraged UK Biobank cognitive tests and vascular morphological features demonstrate significant promise and effectiveness of LAVA in identifying AD stages across the progression continuum.

In situ activation of helper T cells in the lung

To better understand the lung and systemic responses of helper T cells mediating memory immunity to Mycobacterium tuberculosis, we used three- and four-color flow cytometry to study the surface phenotype of CD4(+) lymphocytes. Bronchoalveolar lavage (BAL) fluid and peripheral blood (PB) samples were obtained from a total of 25 subjects, including 10 tuberculosis (TB)-infected subjects, 8 purified-protein-derivative-negative subjects, and 7 purified-protein-derivative-positive subjects. In marked contrast to CD4(+) lymphocytes from PB (9% +/- 5% expressing CD45RA and CD29), the majority (55% +/- 16%) of CD4(+) lymphocytes in BAL (ALs) simultaneously expressed CD45RA, a naïve T-cell marker, and CD29, members of the very late activation family. Further evaluation revealed that CD4(+) ALs expressed both CD45RA and CD45RO, a memory T-cell marker. In addition, the proportion of CD4(+) lymphocytes expressing CD69, an early activation marker, was drastically increased in BAL fluid (83% +/- 9%) compared to PB (1% +/- 1%), whereas no significant difference was seen in the expression of CD25, the low-affinity interleukin 2 receptor (34% +/- 15% versus 40% +/- 16%). More importantly, we identified a minor population of CD69(bright) CD25(bright) CD4(+) lymphocytes in BAL (10% +/- 6%) that were consistently absent from PB (1% +/- 1%). Thus, CD4(+) lymphocytes in the lung paradoxically coexpress surface molecules characteristic of naïve and memory helper T cells as well as surface molecules commonly associated with early and late stages of activation. No difference was observed for ALs obtained from TB-infected and uninfected lung segments in this regard. It remains to be determined if these surface molecules are induced by the alveolar environment or if CD4(+) lymphocytes coexpressing this unusual combination of surface molecules are selectively recruited from the circulation. Our data suggest that ex vivo experiments on helper T-cell subsets that display distinctive phenotypes may be pivotal to studies on the human immune response to potential TB vaccines.