Sociedade Brasileira Infectologia

Pensiun atau purna tugas adalah sebuah peristiwa yang wajar untuk setiap pegawai dari tingkat bawah, menengah hingga atas. Fase ini merupakan babak baru untuk setiap pegawai, sebuah transisi dalam kehidupan yang memerlukan persiapan tersendiri dalam menghadapi kehidupan setelah berakhirnya masa kerja. Seseorang yang telah memasuki masa pensiun akan mendapatkan penghasilan yang jauh lebih sedikit dibandingkan ketika aktif sebagai karyawan, maka dari itu seseorang yang menghadapi masa pensiun memerlukan pengetahuan yang memadai tentang pengelolaan keuangan. Pengetahuan tentang pengelolaan keuangan akan memudahkan seseorang untuk membagi penghasilannya untuk tabungan, investasi, hobi, wirausaha dan lainnya. Pelatihan ini bertujuan untuk membekali para peserta pelatihan yang hendak memasuki masa pensiun dengan pengetahuan tentang pengelolaan keuangan sehingga memudahkan peserta untuk membuat perencanaan dan alokasi keuangan dengan sumber daya yang terbatas setelah memasuki masa pensiun. Pelatihan ini membagi materi pengelolaan keuangan menjadi empat pokok bahasan yaitu alokasi keuangan pasca pensiun, perencanaan investasi, persiapan wirausaha, penyusunan keuangan bisnis secara praktis. Kata Kunci: Manajemen Keuangan, Investasi, Wirausaha, Laporan Keuangan.

Recent advancements in biomedical NLP have been driven by domain-specific pre-trained language models (LMs), yet the challenge of effectively storing extensive biomedical factual knowledge remains. Despite the superior performance of fine-tuned LMs in downstream NLP tasks, these models exhibit limitations in ontology memorization, reasoning abilities, and capturing complex specialized domain terminology. To address these issues, we present four research questions that explore the integration of LMs with large knowledge graphs (KGs) like the Unified Medical Language System (UMLS). Our proposal introduces novel alignment methods to bridge LMs with the UMLS KG, with the aim of leveraging structured background knowledge to enhance the reasoning and generalization capabilities of biomedical LMs. The research proposal discusses multilingual specifics of KBs and evaluation metrics across various datasets.

Contemporary dance is one part of the dance arts in West Java, especially in Bandung city. The contemporary dance in this work raises the issue of communication behavior which is the identity of the Sundanese, namely Someah. The creation of this Someah-themed contemporary dance is a form of collaborative creation of innovative products from the collaboration of culture and technology. Introducing dance art through conventional media in the form of books is starting to become less attractive, especially among the younger generations. This work is expected to be an alternative solution. The study uses the theory of Chris Atton’s Alternative Media (2002). Creating this work also applies the 4P method as an innovation phase which includes person, product, process and press. The research conducted in the process of creating the work includes selecting media for scanning objects, taking pictures and forming Augmented Reality main media. The result of the study of the innovation process as a form of cultural engineering is Someah-themed dance which is published in Augmented Reality as an alternative media for introducing dance. Producing Augmented Reality is considered as a medium to bridge between products and cultural values so that they can synergize in the midst of technological developments.Keywords: Dance, Someah, Alternative Media, Augmented RealityABSTRAKTari kontemporer adalah salah satu bagian dari seni tari di Jawa Barat khususnya di kota Bandung. Tari kontemporer pada karya ini mengangkat tentang prilaku komunikasi yang menjadi identitas dari suku Sunda yaitu Someah. Penciptaan tari kontemporer bertema Someah ini adalah wujud kolaborasi penciptaan produk inovasi dari kolaborasi kebudayaan dan teknologi. Pengenalan seni tari melalui media konvensional berupa buku mulai kurang diminati terutama di kalangan generasi muda. Karya ini diharapkan dapat menjadi solusi alternatif. Kajian ini menggunakan teori Alternatif Media dari Chris Atton (2002). Pembuatan karya ini juga menggunakan metode 4P sebagai fase inovasi yang meliputi person, product, process and press. Pembuatan karya ini melalui riset, pemilihan media untuk objek pemindai, pengambilan gambar dan pembentukan media utama Augmented Reality. Hasil kajian karya utama dari proses inovasi sebagai bentuk rekayasa budaya ini adalah Tari bertema Someah yang dimuat dalam bentuk Augmented Reality sebagai salah satu alternatif media pengenalan tari. Pembuatan Augmented Reality ini juga dipilih sebagai media untuk menjembatani produk serta nilai kebudayaan agar dapat bersinergi ditengah perkembangan teknologi.Kata Kunci: Tari, Someah, Alternatif Media, Augmented Reality

We introduce two data augmentation techniques, which, used with a Resnet-BiLSTM-CTC network, significantly reduce Word Error Rate and Character Error Rate beyond best-reported results on handwriting text recognition tasks. We apply a novel augmentation that simulates strikethrough text (HandWritten Blots) and a handwritten text generation method based on printed text (StackMix), which proved to be very effective in handwriting text recognition tasks. StackMix uses weakly-supervised framework to get character boundaries. Because these data augmentation techniques are independent of the network used, they could also be applied to enhance the performance of other networks and approaches to handwriting text recognition. Extensive experiments on ten handwritten text datasets show that HandWritten Blots augmentation and StackMix significantly improve the quality of handwriting text recognition models.

Aggregating knowledge about drug, disease, and drug reaction entities across a broader range of domains and languages is critical for information extraction (IE) applications. In this work, we present a fine-grained evaluation intended to understand the efficiency of multilingual BERT-based models for biomedical named entity recognition (NER) and multi-label sentence classification tasks. We investigate the role of transfer learning (TL) strategies between two English corpora and a novel annotated corpus of Russian reviews about drug therapy. Labels for sentences include health-related issues or their absence. The sentences with one are additionally labelled at the expression level to identify fine-grained subtypes such as drug names, drug indications, and drug reactions. Evaluation results demonstrate that BERT trained on Russian and English raw reviews (5M in total) shows the best transfer capabilities on evaluation of adverse drug reactions on Russian data. The macro F1 score of 74.85% in the NER task was achieved by our RuDR-BERT model. For the classification task, our EnRuDR-BERT model achieves the macro F1 score of 70%, gaining 8.64% over the score of a general domain BERT model.

This work provides a rigorous detailed kinetic study on the C2H2 + NH2 reaction in a wide range of conditions (T = 250-2000 K & P = 1-76000 Torr). In particular, the composite method W1U was used to construct the potential energy surface on which the kinetic behaviors were characterized within the state-of-the-art master equation/Rice-Ramsperger-Kassel-Marcus (ME/RRKM) framework. Corrections of the hindered internal rotation (HIR) treatment and quantum tunneling effect were included. A clear reaction mechanism shift with respect to both temperature and pressure was revealed via detailed kinetic and species analyses. In particular, bimolecular products (i.e., CH2[double bond, length as m-dash]C[double bond, length as m-dash]NH + H, CH[triple bond, length as m-dash]CNH2 + H, CH3CN + H, CH[triple bond, length as m-dash]C· + NH3 in the decreasing mole fraction order) can be formed directly from the reactants at high temperature and/or low pressure while they can be produced indirectly via intermediates (e.g., ·CH[double bond, length as m-dash]CHNH2(cis), ·CH[double bond, length as m-dash]CHNH2(trans), CH2[double bond, length as m-dash]C·NH2,…) at low temperature and/or high pressure. The calculated rate constants are in good agreement with the literature data from ab initio calculations without any adjustment; thus, the proposed temperature- and pressure-dependent rate constants, together with the thermodynamic data of the species involved, can be confidently used for modeling NH2-related systems under atmospheric and combustion conditions.

Sepsis is a very frequent condition and causes high mortality rates and healthcare costs; it is the main cause of death in intensive care units. Clear, improved prognosis was shown for early diagnosed and treated patients. Treatment guidelines are fundamental for appropriate therapy. It is clear that hypoperfusion patients should be hemodynamically optimized; therefore, recognition of hypoperfusion signs is one of the main therapeutic steps. This guideline discusses the current literature and available data regarding the evaluation of hemodynamic parameters.