Deep Neural Network-Based Method for Detecting Obstructive Meibomian Gland Dysfunction With in Vivo Laser Confocal Microscopy.

PURPOSE: To evaluate the ability of deep learning (DL) models to detect obstructive meibomian gland dysfunction (MGD) using in vivo laser confocal microscopy images. METHODS: For this study, we included 137 images from 137 individuals with obstructive MGD (mean age, 49.9 ± 17.7 years; 44 men and 93 women) and 84 images from 84 individuals with normal meibomian glands (mean age, 53.3 ± 19.6 years; 29 men and 55 women). We constructed and trained 9 different network structures and used single and ensemble DL models and calculated the area under the curve, sensitivity, and specificity to compare the diagnostic abilities of the DL. RESULTS: For the single DL model (the highest model; DenseNet-201), the area under the curve, sensitivity, and specificity for diagnosing obstructive MGD were 0.966%, 94.2%, and 82.1%, respectively, and for the ensemble DL model (the highest ensemble model; VGG16, DenseNet-169, DenseNet-201, and InceptionV3), 0.981%, 92.1%, and 98.8%, respectively. CONCLUSIONS: Our network combining DL and in vivo laser confocal microscopy learned to differentiate between images of healthy meibomian glands and images of obstructive MGD with a high level of accuracy that may allow for automatic obstructive MGD diagnoses in patients in the future.

Hypothermia-induced tyrosine phosphorylation of SIRPα in the brain.

Signal regulatory protein α (SIRPα) is a neuronal membrane protein that undergoes tyrosine phosphorylation in the brain of mice in response to forced swim (FS) stress in cold water, and this response is implicated in regulation of depression-like behavior in the FS test. We now show that subjection of mice to the FS in warm (37 °C) water does not induce the tyrosine phosphorylation of SIRPα in the brain. The rectal temperature (T(rec) ) of mice was reduced to 27° to 30 °C by performance of the FS for 10 min in cold water, whereas it was not affected by the same treatment in warm water. The level of tyrosine phosphorylation of SIRPα in the brain was increased by administration of ethanol or picrotoxin, starvation, or cooling after anesthesia, all of which also induced hypothermia. Furthermore, the tyrosine phosphorylation of SIRPα in cultured hippocampal neurons was induced by lowering the temperature of the culture medium. CD47, a ligand of SIRPα, as well as Src family kinases or SH2 domain-containing protein phosphatase 2 (Shp2), might be important for the basal and the hypothermia-induced tyrosine phosphorylation of SIRPα. Hypothermia is therefore likely an important determinant of both the behavioral immobility and tyrosine phosphorylation of SIRPα observed in the FS test.

Stress-evoked tyrosine phosphorylation of signal regulatory protein α regulates behavioral immobility in the forced swim test.

Severe stress induces changes in neuronal function that are implicated in stress-related disorders such as depression. The molecular mechanisms underlying the response of the brain to stress remain primarily unknown, however. Signal regulatory protein alpha (SIRPalpha) is an Ig-superfamily protein that undergoes tyrosine phosphorylation and binds the protein tyrosine phosphatase Shp2. Here we show that mice expressing a form of SIRPalpha that lacks most of the cytoplasmic region manifest prolonged immobility (depression-like behavior) in the forced swim (FS) test. FS stress induced marked tyrosine phosphorylation of SIRPalpha in the brain of wild-type mice through activation of Src family kinases. The SIRPalpha ligand CD47 was important for such SIRPalpha phosphorylation, and CD47-deficient mice also manifested prolonged immobility in the FS test. Moreover, FS stress-induced tyrosine phosphorylation of both the NR2B subunit of the NMDA subtype of glutamate receptor and the K+-channel subunit Kvbeta2 was regulated by SIRPalpha. Thus, tyrosine phosphorylation of SIRPalpha is important for regulation of depression-like behavior in the response of the brain to stress.