Publications by authors named "Mengyue Lv"
Multi-objective data enhancement for deep learning-based ultrasound analysis.
BMC Bioinformatics
October 2022
Article Synopsis
- Deep Learning is gaining attention for creating automatic treatment recommendations, but small medical datasets can lead to issues like over-fitting.
- The paper introduces a multi-objective data enhancement method that scales up medical data by defining a main task and several auxiliary tasks, allowing models to learn different aspects of the limited data.
- Using a Soft Parameter Sharing approach, the method facilitates knowledge sharing among models, which improves performance; experiments on a thyroid nodule ultrasound dataset demonstrated its effectiveness compared to existing methods.
Following a chronic insult, renal tubular epithelial cells (TECs) contribute to the development of kidney fibrosis through dysregulated lipid metabolism that lead to lipid accumulation and lipotoxicity. Intracellular lipid metabolism is tightly controlled by fatty acids (FAs) uptake, oxidation, lipogenesis, and lipolysis. Although it is widely accepted that impaired fatty acids oxidation (FAO) play a crucial role in renal fibrosis progression, other lipid metabolic pathways, especially FAs uptake, has not been investigated in fibrotic kidney.
View Article and Find Full Text PDFMore than 150 cases of Fanconi syndrome (FS) or hypophosphatemia osteomalacia induced by low-dose adefovir dipivoxil (ADV) have been reported since 2002, when ADV was introduced for the long-term treatment of hepatitis B virus (HBV) infection. Because the life expectancy of HBV-infected individuals has increased, the adverse effects of long-term treatment with antiviral therapies are increasingly observed, and nephrotoxicity is one of the most severe adverse effects of ADV. Therefore, the number of cases may be far higher than reported.
View Article and Find Full Text PDF