TweepFake: About detecting deepfake tweets.

The recent advances in language modeling significantly improved the generative capabilities of deep neural models: in 2019 OpenAI released GPT-2, a pre-trained language model that can autonomously generate coherent, non-trivial and human-like text samples. Since then, ever more powerful text generative models have been developed. Adversaries can exploit these tremendous generative capabilities to enhance social bots that will have the ability to write plausible deepfake messages, hoping to contaminate public debate. To prevent this, it is crucial to develop deepfake social media messages detection systems. However, to the best of our knowledge no one has ever addressed the detection of machine-generated texts on social networks like Twitter or Facebook. With the aim of helping the research in this detection field, we collected the first dataset of real deepfake tweets, TweepFake. It is real in the sense that each deepfake tweet was actually posted on Twitter. We collected tweets from a total of 23 bots, imitating 17 human accounts. The bots are based on various generation techniques, i.e., Markov Chains, RNN, RNN+Markov, LSTM, GPT-2. We also randomly selected tweets from the humans imitated by the bots to have an overall balanced dataset of 25,572 tweets (half human and half bots generated). The dataset is publicly available on Kaggle. Lastly, we evaluated 13 deepfake text detection methods (based on various state-of-the-art approaches) to both demonstrate the challenges that Tweepfake poses and create a solid baseline of detection techniques. We hope that TweepFake can offer the opportunity to tackle the deepfake detection on social media messages as well.

Does monitoring of pre-/post-dialyzer pressure difference improve efficiency in intermittent hemodialysis?

BACKGROUND: Continuous monitoring of pre-/post-dialyzer pressure difference (DeltaP) is widely used in continuous renal replacement therapies to monitor extracorporeal circuit function. The aim of this study was to verify whether DeltaP may help to identify chronic subclinical worsening of dialysis quality due to incomplete dialyzer clotting in intermittent hemodialysis. METHODS: Nine chronic hemodialysis patients were enrolled in the study and dialyzed twice (high-flux polysulfone dialyzer) with DeltaP and urea-clearance monitoring: the first session with a standard anticoagulation and the second without. To verify whether a visible clotting of the dialyzer precedes or follows a significant DeltaP increase, we checked the dialyzers for the presence of red clots after a saline flush performed when a 50% increase in DeltaP was registered. RESULTS: In the second dialysis session after a 50% increase in DeltaP (documented in 7/9 patients), all dialyzers, after saline flush, showed a visible fiber clotting but not a significant reduction (>15%) in urea clearance. In the majority of the patients (6/7), until a few minutes before complete occlusion of the extracorporeal circuit, the urea clearance did not change significantly (-8.9 +/- 12.7%). CONCLUSIONS: The usual check of the presence or absence of red clots in the dialyzer at the end of the dialysis session is enough, in the absence of red clots, to ensure that dialyzer efficiency is maintained during the whole treatment. Contrary to what is applied in CRRT, a continuous monitoring of DeltaP during intermittent hemodialysis would not significantly help to unmask unnoticed inefficient hemodialysis sessions.