Imaging Informatics Division
Distribution based MIL pooling filters: Experiments on a lymph node metastases dataset
Multiple instance learning (MIL), MIL pooling filters, Distribution pooling, Point estimate based pooling
ReadGlobal Epidemiology and Genetics of Hepatocellular Carcinoma
Keywords: Hepatocellular Carcinoma; Epidemiology; Genetics.
ReadAn AI-assisted tool for efficient prostate cancer diagnosis in low-grade and low-volume cases
Pathologists diagnose prostate cancer by core needle biopsy. In low-grade and low-volume cases, they look for a few malignant glands out of hundreds within a core. They may miss a few malignant glands, resulting in repeat biopsies or missed therapeutic opportunities. This study developed a multi-resolution deep-learning pipeline to assist pathologists in detecting malignant glands in core needle biopsies of low-grade and low-volume cases. Analyzing a gland at multiple resolutions, our model exploited morphology and neighborhood information, which were crucial in prostate gland classification. We developed and tested our pipeline on the slides of a local cohort of 99 patients in Singapore. Besides, we made the images publicly available, becoming the first digital histopathology dataset of patients of Asian ancestry with prostatic carcinoma. Our multi-resolution classification model achieved an area under the receiver operating characteristic curve (AUROC) value of 0.992 (95% confidence interval [CI]: 0.985–0.997) in the external validation study, showing the generalizability of our multi-resolution approach.
ReadComparison of protein extraction protocols and allergen mapping from black soldier fly Hermetia illucens
We demonstrated that the selection of protein extraction buffer and the processing method could influence protein yield and cross-reactive allergen detection from processed and un-processed black soldier fly (BSF) samples. In total, 33 putative allergens were detected by comparing the detected BSF proteins to sequences from public allergen protein databases. An LC-MRM-MS assay was developed for tropomyosin, indicating the importance of buffer selection and processing conditions to reduce BSF samples' allergenicity.
ReadNeuroanatomical subtypes of schizophrenia and relationship with illness duration and deficit status
Schizophrenia (SCZ) is a chronic and serious mental illness affecting >27 million worldwide (GBD 2017 Disease and Injury Incidence and Prevalence Collaborators, 2018). Heterogeneity in neurobiology (Voineskos et al., 2020), polygenic scores (Alnæs et al., 2019), symptom presentations (Van Rheenen et al., 2017), treatment responses (Malhotra, 2015), and outcomes (Huber, 1997) and their inter-relationships remain incompletely understood and can potentially confound evaluation and affect management of the condition (Lakhan and Vieira, 2009). Grouping patients by clinical subtypes allows for a deeper examination of homogenous groups, and this is often done using measures which assess symptomatology (e.g positive/negative symptom domains), cognitive status or illness course (e.g remission status) (Tan et al., 2020; Weinberg et al., 2016; Wong et al., 2020). Previous studies have attempted to link these clinical subtypes in SCZ based on deficit syndrome (Tan et al., 2020), cognitive functioning (Ho et al., 2020; Seaton et al., 2001; Weinberg et al., 2016) and remission status (Wong et al., 2020) with underlying neural substrates.
ReadA bio-functional polymer that prevents retinal scarring through modulation of NRF2 signalling pathway
Aberrant wound healing, which leads to irreversible fibrosis, underpins many pathological diseases. In the eye, this manifests as proliferative vitreoretinopathy (PVR), a major cause of poor vision due to failed retinal detachment (RD) surgery1,2. PVR occurs due to proliferative and contractile fibrocellular scar membranes that form either in the vitreous or surrounding the retina.
ReadTransitional premonocytes emerge in the periphery for host defense against bacterial infections
Circulating Ly6Chi monocytes often undergo cellular death upon exhaustion of their antibacterial effector functions, which limits their capacity for subsequent macrophage differentiation. This shrouds the understanding on how the host replaces the tissue-resident macrophage niche effectively during bacterial invasion to avert infection morbidity.
ReadExplaining adversarial vulnerabiity with a data sparisty hypothesis
Keywords : Adversarial vulnerability, Adversarial attacks, Decision boundary, Regularization, Data Sparsity, Smooth SoftMax score surface
Read