Scanned documents (e.g., faxes) are still widely used in clinical practice and are prevalent in Electronic Health Records (EHR). Unlocking information in scanned documents in EHRs is critical for clinical operation and research. However, it is challenging as it requires converting images to texts before applying information extraction technologies. Here we propose a multi-modal approach (ClinicalLayoutLM) that jointly models text extracted from Optical Character Recognition (OCR) and layout/image information to classify scanned clinical documents into different categories (e.g., lab reports and CT scans). Using a clinical corpus of 348, 311 scanned documents, we continually pretrained ClinicalLayoutLM based on LayoutLMv3, a multi-modal model from the open domain. For the task to classify the scanned clinical documents into 16 categories, ClinicalLayoutLM achieved an F1 score of 0.9051, which outperformed the baseline model (0.8840) that was based on text from OCR only. ClinicalLayoutLM is the first of its kind of multi-modal models for clinical documents and we believe it could benefit other clinical natural language processing (NLP) tasks such as layout analysis, information extraction and so on. The code is available at https://github.com/UTHealth-CCB/ClinicalLayoutLM and the pre-trained model is available upon request.

Phrase mining is one of the fundamental NLP tasks that can have significant impact on the efficacy of many downstream applications.Many supervised and unsupervised phrase mining approaches have been proposed.Some rely on linguistic analyzers, and others are language agnostic.A daunting challenge in this task is to distinguish quality phrases from noise phrases, which tightly coexists with quality phrases in the entire frequency spectrum.Most existing approaches to phrase mining, however, rely on frequency-based statistics, hence suffer from quality loss.In this paper, we propose an unsupervised phrase mining framework, "PhraseScope", which consists of a sequence of filters, namely cohesion, domain, and graph filters, to remove noise phrase.Each filter is responsible for removing noise phrase of particular characteristics.Collectively, our proposed filters are capable of detecting and removing noise phrases effectively while preserving quality phrases.Our results show significant improvement in both recall and precision over state-of-the-art frameworks when tested on three different domains of datasets.

Omer Anjum, Hongyu Gong, Suma Bhat, Wen-Mei Hwu, JinJun Xiong. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

In this article implementation of carrier frequency offset estimate for 20MHz LTE baseband processing is discussed. LTE (Long Term Evolution) is a wireless communication standard that makes use of some innovative techniques to gain very high data rates (>100Mbps). This goal for such a high throughput also imposes design challenges for the industry and academia such as in the case of handheld mobile devices where the power budget is very limited. Implicitly high throughput means we need more computation power and more energy. On the other hand industry is also struggling for a flexible hardware solution, or software defined a radio (SDR), to amortize the huge cost of required hardware changes as the wireless standards have kept evolving. Design innovations are now needed to confront those challenges of low power and flexible design without changing the hardware. The implementation is made on Transport Triggered Architecture (TTA), which is a unique concept in computer architecture design, based on the single instruction, “MOVE”. The power consumption of the architecture when synthesized on 180nm technology at 180MHz and 1.8V is 18.39mW. The total area occupied excluding memory is 0.6mm 2 . The proposed TTA solution has been compared with, a more ASIC (application specific integrated circuits), like ASIP (application specific instruction processor) solution and a coprocessor accelerator-based solution. The proposed solution is more flexible: easily programmable due to high level language support, easily scalable, and still efficient in energy consumption needed to complete the CFO (carrier frequency offset) estimation task. Because of these attractive characteristics, TTA is also a potential candidate for SDR platforms.