Research projects, including those focused on cancer, rely on the manual extraction of information from clinical reports. This process is time-consuming and prone to errors, limiting the efficiency of data-driven approaches in healthcare. To address these challenges, Natural Language Processing (NLP) offers an alternative for automating the extraction of relevant data from electronic health records (EHRs). In this study, we focus on lung and breast cancer due to their high incidence and the significant impact they have on public health. Early detection and effective data management in both types of cancer are crucial for improving patient outcomes. To enhance the accuracy and efficiency of data extraction, we utilized GMV's NLP tool uQuery, which excels at identifying relevant entities in clinical texts and converting them into standardized formats such as SNOMED and OMOP. uQuery not only detects and classifies entities but also associates them with contextual information, including negated entities, temporal aspects, and patient-related details. In this work, we explore the use of NLP techniques, specifically Named Entity Recognition (NER), to automatically identify and extract key clinical information from EHRs related to these two cancers. A dataset from Health Research Institute Hospital La Fe (IIS La Fe), comprising 200 annotated breast cancer and 400 lung cancer reports, was used, with eight clinical entities manually labeled using the Doccano platform. To perform NER, we fine-tuned the bsc-bio-ehr-en3 model, a RoBERTa-based biomedical linguistic model pre-trained in Spanish. Fine-tuning was performed using the Transformers architecture, enabling accurate recognition of clinical entities in these cancer types. Our results demonstrate strong overall performance, particularly in identifying entities like MET and PAT, although challenges remain with less frequent entities like EVOL.

Average calibration of the uncertainties of machine learning regression tasks can be tested in two ways. One way is to estimate the calibration error (CE) as the difference between the mean absolute error (MSE) and the mean variance (MV) or mean squared uncertainty. The alternative is to compare the mean squared z-scores or scaled errors (ZMS) to 1. Both approaches might lead to different conclusion, as illustrated on an ensemble of datasets from the recent machine learning uncertainty quantification literature. It is shown here that the CE is very sensitive to the distribution of uncertainties, and notably to the presence of outlying uncertainties, and that it cannot be used reliably for calibration testing. By contrast, the ZMS statistic does not present this sensitivity issue and offers the most reliable approach in this context. Implications for the validation of conditional calibration are discussed.

Deep learning is the state-of-the-art for medical imaging tasks, but requires large, labeled datasets. For risk prediction (e.g., predicting risk of future cancer), large datasets are rare since they require both imaging and long-term follow-up. However, the release of publicly available imaging data with diagnostic labels presents an opportunity for self and semi-supervised approaches to use diagnostic labels to improve label efficiency for risk prediction. Though several studies have compared self-supervised approaches in natural image classification, object detection, and medical image interpretation, there is limited data on which approaches learn robust representations for risk prediction. We present a comparison of semi-and self-supervised learning to predict mortality risk using chest x-ray images. We find that a semi-supervised autoencoder outperforms contrastive and transfer learning in internal and external validation data.

In nuclear engineering, modeling and simulations (M&Ss) are widely applied to support risk-informed safety analysis. Since nuclear safety analysis has important implications, a convincing validation process is needed to assess simulation adequacy, i.e., the degree to which M&S tools can adequately represent the system quantities of interest. However, due to data gaps, validation becomes a decision-making process under uncertainties. Expert knowledge and judgments are required to collect, choose, characterize, and integrate evidence toward the final adequacy decision. However, in validation frameworks CSAU: Code Scaling, Applicability, and Uncertainty (NUREG/CR-5249) and EMDAP: Evaluation Model Development and Assessment Process (RG 1.203), such a decision-making process is largely implicit and obscure. When scenarios are complex, knowledge biases and unreliable judgments can be overlooked, which could increase uncertainty in the simulation adequacy result and the corresponding risks. Therefore, a framework is required to formalize the decision-making process for simulation adequacy in a practical, transparent, and consistent manner. This paper suggests a framework "Predictive Capability Maturity Quantification using Bayesian network (PCMQBN)" as a quantified framework for assessing simulation adequacy based on information collected from validation activities. A case study is prepared for evaluating the adequacy of a Smoothed Particle Hydrodynamic simulation in predicting the hydrodynamic forces onto static structures during an external flooding scenario. Comparing to the qualitative and implicit adequacy assessment, PCMQBN is able to improve confidence in the simulation adequacy result and to reduce expected loss in the risk-informed safety analysis.

Nowadays, AI state-of-the-art techniques includes, among other things, comparing multiple machine learning models within one tool/library. However, it's still very common to try different techniques manually, which – when done over and over – is both boring and prone to mistakes. Diminishing risk of failures gets even more important when having large amounts of code, as debugging and testing gets laborious. With this in mind, AI-JACK was created. It's a machine learning pipeline accelerator, designed as R library to solve problems quicker and better.

Coronavirus disease (COVID-19) spread forecasting is an important task to track the growth of the pandemic. Existing predictions are merely based on qualitative analyses and mathematical modeling. The use of available big data with machine learning is still limited in COVID-19 growth prediction even though the availability of data is abundance. To make use of big data in the prediction using deep learning, we use long short-term memory (LSTM) method to learn the correlation of COVID-19 growth over time. The structure of an LSTM layer is searched heuristically until the best validation score is achieved. First, we trained training data containing confirmed cases from around the globe. We achieved favorable performance compared with that of the recurrent neural network (RNN) method with a comparable low validation error. The evaluation is conducted based on graph visualization and root mean squared error (RMSE). We found that it is not easy to achieve the same quantity of confirmed cases over time. However, LSTM provide a similar pattern between the actual cases and prediction. In the future, our proposed prediction can be used for anticipating forthcoming pandemics. The code is provided here: https://github.com/cbasemaster/lstmcorona

Slides of our EMNLP 2019 talk are avialable here. All the results in the table are produced exactly with this code base. Since VQA and GQA test servers only allow limited number of'Test-Standard' submissions, we use our remaining submission entry from the VQA/GQA challenges 2019 to get these results. For NLVR2, we only test once on the unpublished test set (test-U). We use this code (with model ensemble) to participate in VQA 2019 and GQA 2019 challenge in May 2019. We are the only team ranking top-3 in both challenges.

With the right process in place, it will not be difficult to find state-of-the-art hyperparameter configuration for a given prediction task. Out of the three approaches -- manual, machine-assisted, and algorithmic -- this article will focus on machine-assisted. The article will cover how I do it, get to the proof that the method works, and provide the understanding of why it works. The main principle is simplicity. The first point about performance relates to the issue of accuracy (and other more robust metrics) as a way to measure model performance.