Code Hierarchy Research Articles

Using Enhanced Representations to Predict Medical Procedures from Clinician Notes

Nowadays, most health professionals use electronic health records to keep track of patients. To properly use and share these data, the community has relied on medical classification standards to represent patient information. However, the coding process is tedious and time-consuming, often limiting its application. This paper proposes a novel feature representation method that considers the distinction between diagnoses and procedure codes, and applies this to the task of medical procedure code prediction. Diagnosis codes are combined with text annotations, and the result is then used as input to a downstream procedure code prediction task. Various diagnosis code representations are considered by exploiting a code hierarchy. Furthermore, different text representation strategies are also used, including embeddings from language models. Finally, the method was evaluated using the MIMIC-III database. Our experiments showed improved performance in procedure code prediction when exploiting the diagnosis codes, outperforming state-of-the-art models.

Attention Mechanisms in Clinical Text Classification: A Comparative Evaluation.

Attention mechanisms are now a mainstay architecture in neural networks and improve the performance of biomedical text classification tasks. In particular, models that perform automated medical encoding of clinical documents make extensive use of the label-wise attention mechanism. A label-wise attention mechanism increases a model's discriminatory ability by using label-specific reference information. This information can either be implicitly learned during training or explicitly provided through embedded textual code descriptions or information on the code hierarchy; however, contemporary studies arbitrarily select the type of label-specific reference information. To address this shortcoming, we evaluated label-wise attention initialized with either implicit or explicit label-specific reference information against two common baseline methods-target-attention and text-encoder architecture-specific methods-to generate document embeddings across four text-encoder architectures-a convolutional neural network, two recurrent neural networks, and a transformer. We also present an extension of label-wise attention that can embed the information on the code hierarchy. We performed our experiments on the MIMIC III dataset, which is a standard dataset in the clinical text classification domain. Our experiments showed that using pretrained reference information and the hierarchical design helped improve classification performance. These performance improvements had less impact on larger datasets and label spaces across all text-encoder architectures. In our analysis, we used an attention mechanism's energy scores to explain the perceived differences in performance and interpretability between the text-encoder architectures and types of label-attention.

Hyperbolic graph convolutional neural network with contrastive learning for automated ICD coding

The International Classification of Diseases (ICD) is a widely used criterion for disease classification, health monitoring, and medical data analysis. Deep learning-based automated ICD coding has gained attention due to the time-consuming and costly nature of manual coding. The main challenges of automated ICD coding include imbalanced label distribution, code hierarchy and noisy texts. Recent works have considered using code hierarchy or description for better label representation to solve the problem of imbalanced label distribution. However, these methods are still ineffective and redundant since they only interact with a constant label representation. In this work, we introduce a novel Hyperbolic Graph Convolutional Network with Contrastive Learning (HGCN-CL) to solve the above problems and the shortcomings of the previous methods. We adopt a Hyperbolic graph convolutional network on ICD coding to capture the hierarchical structure of codes, which can solve the problem of large distortions when embedding hierarchical structure with graph convolutional network. Besides, we introduce contrastive learning for automatic ICD coding by injecting code features into text encoder to generate hierarchical-aware positive samples to solve the problem of interacting with constant code features. We conduct experiments on the public MIMIC-III and MIMIC-II datasets. The results on MIMIC III show that HGCN-CL outperforms previous state-of-art methods for automatic ICD coding, which achieves a 2.7% and 3.6% improvement respectively compared to previous best results (Hypercore). We also provide ablation experiments and hierarchy visualization to verify the effectiveness of components in our model.

The b-symbol weight hierarchy of the Kasami codes

The symbol-pair read channel was first proposed by Cassuto and Blaum. Later, Yaakobi et al. generalized it to the b-symbol read channel. It is motivated by the limitations of the reading process in high density data storage systems. One main task in b-symbol coding theory is to determine the b-symbol weight hierarchy of codes. In this paper, we study the b-symbol weight hierarchy of the Kasami codes, which are well known for their applications to construct sequences with optimal correlation magnitudes. The complete symbol-pair weight distribution of the Kasami codes is determined.

Generalized Hamming Weights of Linear Codes From Quadratic Forms Over Finite Fields of Even Characteristic

The generalized Hamming weight of linear codes is a natural generalization of the minimum Hamming distance. They convey the structural information of a linear code and determine its performance in various applications, and have become one of important research topics in coding theory. Recently, Li (IEEE Trans. Inf. Theory, 67(1): 124-129, 2021) and Li and Li (Discrete Math., 345: 112718, 2022) obtained the complete weight hierarchy of linear codes from quadratic forms over finite fields of odd characteristic by analysis of the solutions of the restricted quadratic equation in its subspace. In this paper, we further determine the complete weight hierarchy of linear codes from quadratic forms over finite fields of even characteristic by carefully studying the behavior of the corresponding restricted quadratic forms to the subspaces of the field, and complement the results of Li and Li.

Generalized Weights of Convolutional Codes

In 1997 Rosenthal and York defined generalized Hamming weights for convolutional codes, by regarding a convolutional code as an infinite dimensional linear code endowed with the Hamming metric. In this paper, we propose a new definition of generalized weights of convolutional codes, that takes into account the underlying module structure of the code. We derive the basic properties of our generalized weights and discuss the relation with the previous definition. We establish upper bounds on the weight hierarchy of MDS and MDP codes and show that, depending on the code parameters, some or all of the generalized weights of MDS codes are determined by the length, rank, and internal degree of the code. We also prove an anticode bound for convolutional codes and define optimal anticodes as the codes which meet the anticode bound. Finally, we classify optimal anticodes and compute their weight hierarchy.

0610 Development of a Rule-Based Text Mining Algorithm to Identify Sleep Complaints in Primary Care Progress Notes

Abstract Introduction Sleep complaints are among the most common reasons to seek medical attention, yet sleep disorders are largely underdiagnosed in primary care settings. The ability to process large collections of unstructured clinical notes might offer an opportunity to promote screening of patients suffering from significant sleep disorders. The goal of this study was to develop a simple rule-based algorithm to identify sleep complaints in progress notes from primary care encounters and validate the performance of the algorithm against manual chart review. Methods De-identified progress notes of a random sample of patients with primary care encounters at the University of Kansas Health System in 2019 were extracted from the institution’s clinical research data warehouse (Healthcare Enterprise Repository for Ontological Narration). Review of 163 notes from patients enriched for presence (N=95) or absence (N=68) of sleep disorders based on the International Classification of Disease (ICD)-10 code hierarchy G47 guided the development of a vocabulary of sleep complaints and symptoms, including corresponding negation terms. The vocabulary was used to design a rule-based, regular expression matching algorithm, which was evaluated against manual chart review of the same patient cohort (training dataset). An independent set of notes from another sample of patients with primary care encounters (N=77; testing dataset) was also manually reviewed and used to assess the validation performance of the algorithm. Results In the training dataset, the algorithm had a sensitivity=75%; specificity=91%, positive predictive value (PPV)=90%, and a negative predictive value (NPV)=87%. The area under the receiver operating characteristics (AUC) curve in the training set was 0.84. When the algorithm was evaluated in the testing dataset, we found a sensitivity=53%, specificity=91%, PPV=78%, and NPV=77%. The AUC in the testing dataset was 0.78. Conclusion A simple pattern matching algorithm designed to identify sleep complaints in primary care progress notes showed good performance in the training set and acceptable performance in the testing set. Further refinement of this algorithm with potential incorporation of natural language processing might offer a feasible approach to screen patients for underdiagnosed sleep disorders using primary care clinical notes. Support (If Any) NIH CTSA UL1TR002366.

Summarizing source code with hierarchical code representation

Code summarization aims to automatically generate natural language descriptions for code, and has become a rapidly expanding research area. Data-driven code summarization models based on neural networks have proliferated in recent few years. Almost all of existing neural models are built upon the granularity of token or AST node. This has several drawbacks: a) Code summarization requires high-level knowledge of code while token representations are limited to provide a global view; b) Such approaches can hardly model the hierarchy of code; c) Long input codes challenge such models to handle long-range dependencies due to the large number of tokens and AST nodes. To address these issues, we propose a novel framework to utilize hierarchical representation of code to generate better summaries. We consider two levels of code hierarchy: token-level and statement-level. Our framework contains a pair of customized encoder-decoder models for tokens and AST of code respectively. Each of them has a hierarchical encoder that aims to extract both token and statement-level code features, and an attentional decoder with the ability to attend to those different levels of representation during decoding. They are then combined to predict summaries via ensemble learning. We conduct extensive experiments to evaluate our models on a large Java corpus. The experimental results show that our approach outperforms several state-of-the-art baselines by a substantial margin. In conclusion, our approach could better learn global information of code and shift attention between important statements during summary generation. With the help of hierarchical attention, the models are able to locate keywords more accurately in a top-down way. Ensemble learning is also proved to be an effective way to benefit from multiple input sources.

Perspectives on recovery from older adult trauma survivors living in rural areas

BackgroundOlder patients living in rural areas face unique challenges after trauma that may hinder optimal recovery. This study aims to qualitatively assess postdischarge challenges in this vulnerable population.MethodsWe conducted remote...

Hierarchy of Cultural Codes in Philosophical and Existential Interpretation of Cross-Cultural Media Production

Hierarchy of Cultural Codes in Philosophical and Existential Interpretation of Cross-Cultural Media Production

Weight distributions and weight hierarchies of a family of p-ary linear codes

The weight distribution and weight hierarchy of linear codes are two important research topics in coding theory. In this paper, by choosing proper defining sets from inhomogeneous quadratic functions over $\mathbb{F}_{q}^{2},$ we construct a family of $3$-weight $p$-ary linear codes and determine their weight distributions and weight hierarchies. Most of the codes can be used in secret sharing schemes.

Encoding Diagnostic Codes for Privacy-Preserving Record Linkage

IntroductionDiagnostic codes, such as the ICD-10, may be considered as sensitive information. If such codes have to be encoded using current methods for data linkage, all hierarchical information given by the code positions will be lost. We present a technique (HPBFs) for preserving the hierarchical information of the codes while protecting privacy. The new method modifies a widely used Privacy-preserving Record Linkage (PPRL) technique based on Bloom filters for the use with hierarchical codes.
 Objectives and ApproachAssessing the similarities of hierarchical codes requires considering the code positions of two codes in a given diagnostic hierarchy. The hierarchical similarities of the original diagnostic code pairs should correspond closely to the similarity of the encoded pairs of the same code.
 Furthermore, to assess the hierarchy-preserving properties of an encoding, the impact on similarity measures from differing code positions at all levels of the code hierarchy can be evaluated. A full match of codes should yield a higher similarity than partial matches.
 Finally, the new method is tested against ad-hoc solutions as an addition to a standard PPRL setup. This is done using real-world mortality data with a known link status of two databases.
 ResultsIn all applications for encoded ICD codes where either categorical discrimination, relational similarity or linkage quality in a PPRL setting is required, HPBFs outperform other known methods. Lower mean differences and smaller confidence intervals between clear-text codes and encrypted code pairs were observed, indicating better preservation of hierarchical similarities. Finally, using these techniques allows for much better hierarchical discrimination for partial matches.
 ConclusionThe new technique yields better linkage results than all other known methods to encrypt hierarchical codes. In all tests, comparing categorical discrimination, relational similarity and PPRL linkage quality, HPBFs outperformed methods currently used.

MP11: Emergency physician attitudes on opioid use disorder and barriers to providing buprenorphine/naloxone

Introduction: Buprenorphine/naloxone (buprenorphine) has proven to be a life-saving intervention amidst the ongoing opioid epidemic in Canada. Research has shown benefits to initiating buprenorphine from the emergency department (ED) including improved treatment retention, systemic health care savings and fewer drug-related visits to the ED. Despite this, there has been little to no uptake of this evidence-based practice in our department. This qualitative study aimed to determine the local barriers and potential solutions to initiating buprenorphine in the ED and gain an understanding of physician attitudes and behaviours regarding harm reduction care and opioid use disorder management. Methods: ED physicians at a midsize Atlantic hospital were recruited by convenience sampling to participate in semi-structured privately conducted interviews. Audio recordings were transcribed verbatim and de-identified transcripts were uploaded to NVivo 12 plus for concept driven and inductive coding and a hierarchy of open, axial and selective coding was employed. Transcripts were independently reviewed by a local qualitative research expert and themes were compared for similarity to limit bias. Interview saturation was reached after 7 interviews. Results: Emergent themes included a narrow scope of harm reduction care that primarily focused on abstinence-based therapies and a multitude of biases including feelings of deception, fear of diversion, feeling buprenorphine induction was too time consuming for the ED and differentiating patients with opioid use disorder from ‘medically ill’ patients. Several barriers and proposed solutions to initiating buprenorphine from the ED were elicited including lack of training and need for formal education, poor familiarity with buprenorphine, the need for an algorithm and community bridge program and formal supports such as an addictions consult team for the ED. Conclusion: This study elicited several opportunities for improved care for patients with addictions presenting to our ED. Future education will focus on harm reduction care, specifically strategies for managing patients desiring to continue to use substances. Education will focus on addressing the multitude of biases elicited and dispelling common myths. A locally informed buprenorphine pathway will be developed. In future, this study may be used to advocate for improved formal supports for our department including an addictions consult team.

Reinforcement-Learning-Guided Source Code Summarization Using Hierarchical Attention

Code summarization (aka comment generation) provides a high-level natural language description of the function performed by code, which can benefit the software maintenance, code categorization and retrieval. To the best of our knowledge, the state-of-the-art approaches follow an encoder-decoder framework which encodes source code into a hidden space and later decodes it into a natural language space. Such approaches suffer from the following drawbacks: (a) they are mainly input by representing code as a sequence of tokens while ignoring code hierarchy; (b) most of the encoders only input simple features (e.g., tokens) while ignoring the features that can help capture the correlations between comments and code; (c) the decoders are typically trained to predict subsequent words by maximizing the likelihood of subsequent ground truth words, while in real world, they are excepted to generate the entire word sequence from scratch. As a result, such drawbacks lead to inferior and inconsistent comment generation accuracy. To address the above limitations, this paper presents a new code summarization approach using hierarchical attention network by incorporating multiple code features, including type-augmented abstract syntax trees and program control flows. Such features, along with plain code sequences, are injected into a deep reinforcement learning (DRL) framework (e.g., actor-critic network) for comment generation. Our approach assigns weights (pays “attention”) to tokens and statements when constructing the code representation to reflect the hierarchical code structure under different contexts regarding code features (e.g., control flows and abstract syntax trees). Our reinforcement learning mechanism further strengthens the prediction results through the actor network and the critic network, where the actor network provides the confidence of predicting subsequent words based on the current state, and the critic network computes the reward values of all the possible extensions of the current state to provide global guidance for explorations. Eventually, we employ an advantage reward to train both networks and conduct a set of experiments on a real-world dataset. The experimental results demonstrate that our approach outperforms the baselines by around 22 to 45 percent in BLEU-1 and outperforms the state-of-the-art approaches by around 5 to 60 percent in terms of S-BLEU and C-BLEU.

WordCommentsAnalyzer: A windows software tool for qualitative research.

There is a lack of free software that provides a professional and smooth experience in text editing and markup for qualitative data analysis. Word processing software like Microsoft Word provides a good editing experience, allowing the researcher to effortlessly add comments to text portions. However, organizing the keywords and categories in the comments can become a more difficult task when the amount of data increases. We present WordCommentsAnalyzer, a software tool that is written in C# using .NET Framework and OpenXml, which helps a qualitative researcher to organize codes when using Microsoft Word as the primary text markup software. WordCommentsAnalyzer provides an effective user interface to count codes, to organize codes in a code hierarchy, and to see various data extracts belonging to each code. It also offers basic visualization tools. We illustrate how to use this software by conducting a preliminary content analysis on Tweets with the #successfulaging hashtag. We also demonstrate that the software has satisfactory performance on a large dataset of Iranian journals abstracts. We hope this open-source software will facilitate qualitative data analysis by researchers who are interested in using Word for this purpose.

WordCommentsAnalyzer: A windows software tool for qualitative research

There is a lack of free software that provides a professional and smooth experience in text editing and markup for qualitative data analysis. Word processing software like Microsoft Word provides a good editing experience, allowing the researcher to effortlessly add comments to text portions. However, organizing the keywords and categories in the comments can become a more difficult task when the amount of data increases. We present WordCommentsAnalyzer, a software tool that is written in C# using .NET Framework and OpenXml, which helps a qualitative researcher to organize codes when using Microsoft Word as the primary text markup software. WordCommentsAnalyzer provides an effective user interface to count codes, to organize codes in a code hierarchy, and to see various data extracts belonging to each code. It also offers basic visualization tools. We illustrate how to use this software by conducting a preliminary content analysis on Tweets with the #successfulaging hashtag. We also demonstrate that the software has satisfactory performance on a large dataset of Iranian journals abstracts. We hope this open-source software will facilitate qualitative data analysis by researchers who are interested in using Word for this purpose.

WordCommentsAnalyzer: A windows software tool for qualitative research.

There is a lack of free software that provides a professional and smooth experience in text editing and markup for qualitative data analysis. Word processing software like Microsoft Word provides a good editing experience, allowing the researcher to effortlessly add comments to text portions. However, organizing the keywords and categories in the comments can become a more difficult task when the amount of data increases. We present WordCommentsAnalyzer, a software tool that is written in C# using .NET Framework and OpenXml, which helps a qualitative researcher to organize codes when using Microsoft Word as the primary text markup software. WordCommentsAnalyzer provides an effective user interface to count codes, to organize codes in a code hierarchy, and to see various data extracts belonging to each code. We illustrate how to use the software by conducting a preliminary content analysis on Tweets with the #successfulaging hashtag. We hope this open-source software will facilitate qualitative data analysis by researchers who are interested in using Word for this purpose.

Contextualizing spectrum-based fault localization

Context: Fault localization is among the most expensive tasks in software development. Spectrum-based fault localization (SFL) techniques seek to pinpoint faulty program elements (e.g., statements), by sorting them only by their suspiciousness scores. Developers tend to fall back on another debugging strategy if they do not find the bug in the first positions of a suspiciousness list.Objective: In this study, we assess techniques to contextualize code inspection whose goal is two-fold: to provide guidance during fault localization, and to improve the effectiveness of SFL techniques in classifying bugs within the first picks. Code Hierarchy (CH) and Integration Coverage-based Debugging (ICD) techniques provide a search roadmap—a list of methods—that guide the developer toward faults. CH assigns a method with the highest suspiciousness score of its blocks, and ICD captures method call relationships from testing to establish the roadmap. Two new filtering strategies—Fixed Budget (FB) and Level Score (LS)—are combined with ICD and CH for reducing the amount of blocks to inspect in each method.Method: We evaluated the effectiveness of ICD, CH, FB, LS, and a suspiciousness block list (BL) on 62 bugs from 7 real programs.Results: ICD and CH using FB found more faults inspecting less blocks than BL with statistical significance. More than 50% of the faults were found inspecting at most 10 blocks using ICD-FB and CH-FB. Moreover, ICD and CH located 70% of the faults by inspecting, at most, 4 methods.Conclusions: These results suggest that the contextualization provided by roadmaps and filtering strategies is useful for guiding developers toward faults and improves the performance of SFL techniques.

Standard International Trade Classification

Trade classifications are a necessary prerequisite for the compilation of trade statistics, and they should – beyond that – be regarded as a valuable base for the definition of shared controlled vocabularies for linked business data that deal with import, export etc. The Standard International Trade Classification (SITC) provided by the UN Statistics Division is a widely used classification mostly applied for scientific and analytical purposes. SITC – as most other trade classifications – is available today only in text or spreadsheet formats. These formats reveal the inner hierarchical structure of SITC to the human reader, because SITC trade codes are built according to the decimal classification scheme, but unfortunately, SITC’s inner structure is opaque to computer applications in text and spreadsheet formats. The paper discusses an approach to set up an OWL-2 ontology for SITC that states subsumption relations between classes of goods. This kind of semantic underpinning of SITC is suited to ease both checking and extending SITC and to derive from it a shared controlled vocabulary for business linked data. Some problems of today’s SITC (among them missing inner nodes of the trade code hierarchy) are carefully discussed, and the paper motivates several decisions that were taken for ontology design. Finally, the study introduces the semantic reasoner as a tool for the (at least partial) automatic derivation of structural information for SITC from the trade code building rule. The paper reports on reasoner runtimes observed for different versions of the SITC ontology and for different versions of the Pellet reasoner.

CHARACTERIZATIONS FOR SEVERAL CLASSES OF ALTERNATIVE CODES

Alternative codes, an extension of the notion of ordinary codes, have been first introduced and considered by P. T. Huy et al. in 2004. In this paper, we consider some new subclasses of alternative codes. In particular, characteristic properties and a hierarchy of such codes are established. Also, algorithms to test for these classes of codes are proposed.