15 September 2022 in IEEE Access
We propose a text-guided sketch-to-image synthesis model that semantically mixes style and content features from the latent space of an inverted Generative Adversarial Network (GAN). Our goal is to synthesize plausible images from human facial sketches and their respective text descriptions. In our approach, we adapted a generative model termed Contextual GAN (CT-GAN) that efficiently encodes visual-linguistic semantic features pre-trained on over 400 million text-image pairs at different resolutions along the model. Also, we introduced an intermediate mapping network called c-Map that combines textual and visual-based features to a disentangled latent space W + for better feature matching. Furthermore to maximise the computational performance of our model, we implemented a linear-based attention scheme along the pipeline of our model to eliminate the drawbacks of inefficient attention modules that are quadratic in complexity. Finally, the hierarchical setting of our model ensures that textual, style and content features are synthesised based on their unique fine grained details, which result in visually appealing images.
5 April 2025 in The International Journal of Health Planning and Management
Purpose: Post‐cancer treatment pain (PCTP) is prevalent among cancer survivors but remains understudied. It is critical to quantify PCTP prevalence over time and to estimate the resulting short and long‐term incremental healthcare expenditures (total, third‐party, and out‐of‐pocket) and out‐of‐pocket burden among cancer survivors.Methods: A longitudinal retrospective cohort design was used. To identify cancer survivors (≥ 18 years) with PCTP, the Medical Expenditure Panel Survey (MEPS) and its supplementary Cancer Self‐Administered Questionnaire (CSAQ)were used. Recycled predictions from generalised linear models (GLM) with log‐link and gamma distribution were used to estimate annual incremental healthcare expenditures at different PCTP gradations over time. To account for covariate imbalance, sensitivity analysis using inverse probability weighting was conducted.Results: 2125 cancer survivors had PCTP. Post‐cancer treatment, 10.5%–24.2% of survivors experienced some form of chronic PCTP, whereas between 21.9%–5.1% experienced acute PCTP. Across the survivorship journey, the adjusted total annual incremental healthcare expenditures were the highest among cancer survivors with moderate chronic PCTP (< 1‐year post‐cancer treatment), and severe chronic PCTP, (≥ 5 years post‐cancer treatment) compared to survivors with no pain, reaching respectively, $27.3 and $40.2 billion nationally. There was a significant high out‐of‐pocket burden among cancer survivors with severe chronic PCTP compared to those with no pain.Conclusion: These findings highlight the persistent financial burden of PCTP but also the critical need for effective pain management alongside the use of patient‐reported outcomes for pain among cancer survivors.