Bypassing cisplatin resistance in Nrf2 hyperactivated head and neck cancer through effective PI3Kinase targeting.

Background: For patients with head and neck squamous cell carcinoma (HNSCC), failure of definitive radiation combined with cisplatin nearly universally results in death. Although hyperactivation of the Nrf2 pathway can drive radiation and cisplatin resistance along with suppressed anti-tumor immunity, treatment-refractory HNSCC tumors may retain sensitivity to targeted agents secondary to synergistic lethality with other oncogenic drivers (e.g.

Personalized outcomes in neuropathic pain: a clinical relevance and assay sensitivity analysis from a randomized controlled trial.

Objective: To explore the clinical relevance and assay sensitivity of using personalized outcomes using data from a randomized clinical trial (RCT) in people with chemotherapy-induced peripheral neuropathy (CIPN).

Design: This study is a secondary analysis that leveraged data from a RCT of transcutaneous electrical stimulation for CIPN to test whether personalized outcomes could minimize potential floor effects and increase the assay sensitivity of pain clinical trials (ie, ability to detect a true treatment effect).

Setting: Participants were recruited for a RCT from community oncology clinics in the United States.

Towards trustworthy seizure onset detection using workflow notes.

A major barrier to deploying healthcare AI is trustworthiness. One form of trustworthiness is a model's robustness across subgroups: while models may exhibit expert-level performance on aggregate metrics, they often rely on non-causal features, leading to errors in hidden subgroups. To take a step closer towards trustworthy seizure onset detection from EEG, we propose to leverage annotations that are produced by healthcare personnel in routine clinical workflows-which we refer to as workflow notes-that include multiple event descriptions beyond seizures.

A multivariate adaptive gradient algorithm with reduced tuning efforts.

Large neural networks usually perform well for executing machine learning tasks. However, models that achieve state-of-the-art performance involve arbitrarily large number of parameters and therefore their training is very expensive. It is thus desired to implement methods with small per-iteration costs, fast convergence rates, and reduced tuning.