Publications by authors named "Q Ping Dou"

510 Publications

Learning with Privileged Multimodal Knowledge for Unimodal Segmentation.

IEEE Trans Med Imaging 2021 Oct 11;PP. Epub 2021 Oct 11.

Multimodal learning usually requires a complete set of modalities during inference to maintain performance. Although training data can be well-prepared with high-quality multiple modalities, in many cases of clinical practice, only one modality can be acquired and important clinical evaluations have to be made based on the limited single modality information. In this work, we propose a privileged knowledge learning framework with the 'Teacher-Student' architecture, in which the complete multimodal knowledge that is only available in the training data (called privileged information) is transferred from a multimodal teacher network to a unimodal student network, via both a pixel-level and an image-level distillation scheme. Specifically, for the pixel-level distillation, we introduce a regularized knowledge distillation loss which encourages the student to mimic the teacher's softened outputs in a pixel-wise manner and incorporates a regularization factor to reduce the effect of incorrect predictions from the teacher. For the image-level distillation, we propose a contrastive knowledge distillation loss which encodes image-level structured information to enrich the knowledge encoding in combination with the pixel-level distillation. We extensively evaluate our method on two different multi-class segmentation tasks, i.e., cardiac substructure segmentation and brain tumor segmentation. Experimental results on both tasks demonstrate that our privileged knowledge learning is effective in improving unimodal segmentation and outperforms previous methods.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1109/TMI.2021.3119385DOI Listing
October 2021

Anchor-guided online meta adaptation for fast one-Shot instrument segmentation from robotic surgical videos.

Med Image Anal 2021 Sep 20;74:102240. Epub 2021 Sep 20.

Department of Computer Science and Engineering, The Chinese University of Hong Kong, HKSAR, China; T-Stone Robotics Institute, The Chinese University of Hong Kong, HKSAR, China.

The scarcity of annotated surgical data in robot-assisted surgery (RAS) motivates prior works to borrow related domain knowledge to achieve promising segmentation results in surgical images by adaptation. For dense instrument tracking in a robotic surgical video, collecting one initial scene to specify target instruments (or parts of tools) is desirable and feasible during the preoperative preparation. In this paper, we study the challenging one-shot instrument segmentation for robotic surgical videos, in which only the first frame mask of each video is provided at test time, such that the pre-trained model (learned from easily accessible source) can adapt to the target instruments. Straightforward methods transfer the domain knowledge by fine-tuning the model on each given mask. Such one-shot optimization takes hundred of iterations and the test runtime is unfeasible. We present anchor-guided online meta adaptation (AOMA) for this problem. We achieve fast one-shot test time optimization by meta-learning a good model initialization and learning rates from source videos to avoid the laborious and handcrafted fine-tuning. The trainable two components are optimized in a video-specific task space with a matching-aware loss. Furthermore, we design an anchor-guided online adaptation to tackle the performance drop throughout a robotic surgical sequence. The model is continuously adapted on motion-insensitive pseudo-masks supported by anchor matching. AOMA achieves state-of-the-art results on two practical scenarios: (1) general videos to surgical videos, (2) public surgical videos to in-house surgical videos, while reducing the test runtime substantially.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.media.2021.102240DOI Listing
September 2021

Recent Advances in Repurposing Disulfiram and Disulfiram Derivatives as Copper-Dependent Anticancer Agents.

Front Mol Biosci 2021 17;8:741316. Epub 2021 Sep 17.

Departments of Oncology, Pharmacology and Pathology, School of Medicine, Barbara Ann Karmanos Cancer Institute, Wayne State University, Detroit, MI, United States.

Copper (Cu) plays a pivotal role in cancer progression by acting as a co-factor that regulates the activity of many enzymes and structural proteins in cancer cells. Therefore, Cu-based complexes have been investigated as novel anticancer metallodrugs and are considered as a complementary strategy for currently used platinum agents with undesirable general toxicity. Due to the high failure rate and increased cost of new drugs, there is a global drive towards the repositioning of known drugs for cancer treatment in recent years. Disulfiram (DSF) is a first-line antialcoholism drug used in clinics for more than 65 yr. In combination with Cu, it has shown great potential as an anticancer drug by targeting a wide range of cancers. The reaction between DSF and Cu ions forms a copper diethyldithiocarbamate complex (Cu(DDC) also known as CuET) which is the active, potent anticancer ingredient through inhibition of NF-κB and ubiquitin-proteasome system as well as alteration of the intracellular reactive oxygen species (ROS). Importantly, DSF/Cu inhibits several molecular targets related to drug resistance, stemness, angiogenesis and metastasis and is thus considered as a novel strategy for overcoming tumour recurrence and relapse in patients. Despite its excellent anticancer efficacy, DSF has proven unsuccessful in several cancer clinical trials. This is likely due to the poor stability, rapid metabolism and/or short plasma half-life of the currently used oral version of DSF and the inability to form Cu(DDC) at relevant concentrations in tumour tissues. Here, we summarize the scientific rationale, molecular targets, and mechanisms of action of DSF/Cu in cancer cells and the outcomes of oral DSF ± Cu in cancer clinical trials. We will focus on the novel insights on harnessing the immune system and hypoxic microenvironment using DSF/Cu complex and discuss the emerging delivery strategies that can overcome the shortcomings of DSF-based anticancer therapies and provide opportunities for translation of DSF/Cu or its Cu(DDC) complex into cancer therapeutics.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.3389/fmolb.2021.741316DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8484884PMC
September 2021

The epidemiology, pathophysiological mechanisms, and management toward COVID-19 patients with Type 2 diabetes: A systematic review.

Prim Care Diabetes 2021 Sep 6. Epub 2021 Sep 6.

School of Resources and Environmental Science, Wuhan University, Wuhan, China; International Institute of Spatial Lifecourse Epidemiology (ISLE), Wuhan University, Wuhan, China. Electronic address:

This review comprehensively summarizes epidemiologic evidence of COVID-19 in patients with Type 2 diabetes, explores pathophysiological mechanisms, and integrates recommendations and guidelines for patient management. We found that diabetes was a risk factor for diagnosed infection and poor prognosis of COVID-19. Patients with diabetes may be more susceptible to adverse outcomes associated with SARS-CoV-2 infection due to impaired immune function and possible upregulation of enzymes that mediate viral invasion. The chronic inflammation caused by diabetes, coupled with the acute inflammatory reaction caused by SARS-CoV-2, results in a propensity for inflammatory storm. Patients with diabetes should be aware of their increased risk for COVID-19.
View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.pcd.2021.08.014DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8418914PMC
September 2021

Current Status and Future Perspectives on Old Drug Repurposing for Cancer Treatment.

Authors:
Q Ping Dou

Recent Pat Anticancer Drug Discov 2021 ;16(2):120-121

Barbara Ann Karmanos Cancer Institute and Departments of Oncology, Pharmacology and Pathology, School of Medicine Wayne State University, Detroit, MI 48201-2013, United States.

View Article and Find Full Text PDF

Download full-text PDF

Source
http://dx.doi.org/10.2174/157489281602210806102833DOI Listing
January 2021
-->