Athey, J. | bioRxiv Science

bioRxiv2022

Multinuclear MRI reveals early efficacy of stem cell therapy in stroke

Compromised adult human mesenchymal stem cells (hMSC) can impair cell therapy efficacy and further reverse ischemic recovery. However, in vitro assays require extended passage to characterize cells, limiting rapid assessment for therapeutic potency. Multinuclear magnetic resonance imaging and spectroscopy (MRI/S) provides near real-time feedback on disease progression and tissue recovery. Applied to ischemic stroke, 23Na MRI evaluates treatment efficacy within 24 h after middle cerebral artery occlusion, showing recovery of sodium homeostasis and lesion reduction in specimens treated with hMSC while 1H MRS identifies reduction in lactate levels. This combined metric was confirmed by evaluating treatment groups receiving healthy or compromised hMSC versus control (sham saline injection) over 21 d. Behavioral tests to assess functional recovery and cell analysis for immunomodulatory and macrophage activity to detect hMSC potency confirm MR findings. Clinically, these MR metrics may prove critical to early evaluations of therapeutic efficacy and overall stroke recovery.

bioengineering↗

bioRxiv2021

Deep learning from multiple experts improves identification of amyloid neuropathologies

Pathologists can label pathologies differently, making it challenging to yield consistent assessments in the absence of one ground truth. To address this problem, we present a DL approach that draws on a cohort of experts, weighs each contribution, and is robust to noisy labels. We collected 100,495 annotations on 20,099 candidate amyloid beta neuropathologies (cerebral amyloid angiopathy (CAA), and cored and diffuse plaques) from three institutions, independently annotated by five experts. DL methods trained on a consensus-of-two strategy yielded 12.6-26% improvements by area under the precision recall curve (AUPRC) when compared to those that learned individualized annotations. This strategy surpassed individual-expert models, even when unfairly assessed on benchmarks favoring them. Moreover, ensembling over individual models was robust to hidden random annotators. In blind prospective tests of 52,555 subsequent expert-annotated images, the models labeled pathologies like their human counterparts (consensus model AUPRC=0.74 cored; 0.69 CAA). This study demonstrates a means to combine multiple ground truths into a common-ground DL model that yields consistent diagnoses informed by multiple and potentially variable expert opinion.

pathology↗