Use of "off-the-shelf" information extraction algorithms in clinical informatics: A feasibility study of MetaMap annotation of Italian medical notes.

Information extraction from narrative clinical notes is useful for patient care, as well as for secondary use of medical data, for research or clinical purposes. Many studies focused on information extraction from English clinical texts, but less dealt with clinical notes in languages other than English. This study tested the feasibility of using "off the shelf" information extraction algorithms to identify medical concepts from Italian clinical notes. Among all the available and well-established information extraction algorithms, we used MetaMap to map medical concepts to the Unified Medical Language System (UMLS). The study addressed two questions: (Q1) to understand if it would be possible to properly map medical terms found in clinical notes and related to the semantic group of "Disorders" to the Italian UMLS resources; (Q2) to investigate if it would be feasible to use MetaMap as it is to extract these medical concepts from Italian clinical notes. We performed three experiments: in EXP1, we investigated how many medical concepts of the "Disorders" semantic group found in a set of clinical notes written in Italian could be mapped to the UMLS Italian medical sources; in EXP2 we assessed how the different processing steps used by MetaMap, which are English dependent, could be used in Italian texts to map the original clinical notes on the Italian UMLS sources; in EXP3 we automatically translated the clinical notes from Italian to English using Google Translator, and then we used MetaMap to map the translated texts. Results in EXP1 showed that the Italian UMLS Metathesaurus sources covered 91% of the medical terms of the "Disorders" semantic group, as found in the studied dataset. We observed that even if MetaMap was built to analyze texts written in English, most of its processing steps worked properly also with texts written in Italian. MetaMap identified correctly about half of the concepts in the Italian clinical notes. Using MetaMap's annotation on Italian clinical notes instead of a simple text search improved our results of about 15 percentage points. MetaMap's annotation of Italian clinical notes showed recall, precision and F-measure equal to 0.53, 0.98 and 0.69, respectively. Most of the failures were due to the impossibility for MetaMap to generate meaningful variants for the Italian language, suggesting that modifying MetaMap to allow generating Italian variants could improve the performance. MetaMap's performance in annotating automatically translated English clinical notes was in line with findings in the literature, with similar recall (0.75), F-measure (0.83) and even higher precision (0.95). Most of the failures were due to a bad Italian to English translation of medical terms, suggesting that using an automatic translation tool specialized in translating medical concepts might be useful to obtain better performances. In conclusion, performances obtained using MetaMap on the fully automatic translation of the Italian text are good enough to allow to use MetaMap "as it is" in clinical practice.

Implementing standards for the interoperability among healthcare providers in the public regionalized Healthcare Information System of the Lombardy Region.

Information technologies (ITs) have now entered the everyday workflow in a variety of healthcare providers with a certain degree of independence. This independence may be the cause of difficulty in interoperability between information systems and it can be overcome through the implementation and adoption of standards. Here we present the case of the Lombardy Region, in Italy, that has been able, in the last 10 years, to set up the Regional Social and Healthcare Information System, connecting all the healthcare providers within the region, and providing full access to clinical and health-related documents independently from the healthcare organization that generated the document itself. This goal, in a region with almost 10 millions citizens, was achieved through a twofold approach: first, the political and operative push towards the adoption of the Health Level 7 (HL7) standard within single hospitals and, second, providing a technological infrastructure for data sharing based on interoperability specifications recognized at the regional level for messages transmitted from healthcare providers to the central domain. The adoption of such regional interoperability specifications enabled the communication among heterogeneous systems placed in different hospitals in Lombardy. Integrating the Healthcare Enterprise (IHE) integration profiles which refer to HL7 standards are adopted within hospitals for message exchange and for the definition of integration scenarios. The IHE patient administration management (PAM) profile with its different workflows is adopted for patient management, whereas the Scheduled Workflow (SWF), the Laboratory Testing Workflow (LTW), and the Ambulatory Testing Workflow (ATW) are adopted for order management. At present, the system manages 4,700,000 pharmacological e-prescriptions, and 1,700,000 e-prescriptions for laboratory exams per month. It produces, monthly, 490,000 laboratory medical reports, 180,000 radiology medical reports, 180,000 first aid medical reports, and 58,000 discharge summaries. Hence, despite there being still work in progress, the Lombardy Region healthcare system is a fully interoperable social healthcare system connecting patients, healthcare providers, healthcare organizations, and healthcare professionals in a large and heterogeneous territory through the implementation of international health standards.

Challenges of the e-Health curricular education in bio-medical engineering and in medicine.

Curricular recommendations coming from highly respectable associations are highly useful. Nevertheless, they show fatigue in keeping the pace of any fast evolution, as in the ICT happens. So we do the attempt to disclose the emerging challenges affecting e-Health curricular education.

eHealth Turning Points as Forced by the Covid-19 Dramatic Experience.

In this ongoing fall of the year 2021, many disciplines are frightened by the Covid-19 situation. A generalized sense of Scientific and administrative impotence, - in keeping the pandemic under real control, - is felt widely in Society. In this Invited Lecture the author reminds us of the blows suffered, recalls pertinent elements present in our social organization, browses selected eHealth experiences and proposes an open agenda of actions to allow the eHealth to help the population segments better, and individuals as well.

An ontology-based tool for the correspondences between specialist and consumer medical lexicons for the geriatrics domain.

New services devoted to improve personalized healthcare are emerging from information technology developments. Personal health record systems allow the patients to participate actively in their healthcare process. However, the dissemination and use of personal health record systems face with some barriers, for example low health literacy that leads to discrepancy in understanding medical concepts. While it is important to present health information using consumer-familiar terms in consumer applications, consistently converting medical terms to consumer-familiar ones is a challenging task. We designed and developed both an ontology-like taxonomic structure devoted to the Geriatrics domain for the outpatient and a software tool, for carrying out the matching between the medical vocabulary of the consumer and that of the doctor from the outpatient's and their family point of view.

A web-based tool for cooperating behaviors in eating and physical activity control.

The field of information technology and the Internet for health care has developed rapidly in the last few years. Furthermore, new services devoted to improve personalized healthcare are emerging from current web-orientated research. Control of eating and physical activity behaviors can be performed in a computer mediated way as a social networking application. To this purpose, we designed and implemented a web application based on the cooperation between two communities: Patients and Nutritionists. The patients are able to cooperate as within a self-help group, while nutritionists can guide patients struggling with incorrect lifestyle and its consequences.

Statistical analysis of genomic protein family and domain controlled annotations for functional investigation of classified gene lists.

BACKGROUND: The increasing protein family and domain based annotations constitute important information to understand protein functions and gain insight into relations among their codifying genes. To allow analyzing of gene proteomic annotations, we implemented novel modules within GFINDer, a Web system we previously developed that dynamically aggregates functional and phenotypic annotations of user-uploaded gene lists and allows performing their statistical analysis and mining. RESULTS: Exploiting protein information in Pfam and InterPro databanks, we developed and added in GFINDer original modules specifically devoted to the exploration and analysis of functional signatures of gene protein products. They allow annotating numerous user-classified nucleotide sequence identifiers with controlled information on related protein families, domains and functional sites, classifying them according to such protein annotation categories, and statistically analyzing the obtained classifications. In particular, when uploaded nucleotide sequence identifiers are subdivided in classes, the Statistics Protein Families&Domains module allows estimating relevance of Pfam or InterPro controlled annotations for the uploaded genes by highlighting protein signatures significantly more represented within user-defined classes of genes. In addition, the Logistic Regression module allows identifying protein functional signatures that better explain the considered gene classification. CONCLUSION: Novel GFINDer modules provide genomic protein family and domain analyses supporting better functional interpretation of gene classes, for instance defined through statistical and clustering analyses of gene expression results from microarray experiments. They can hence help understanding fundamental biological processes and complex cellular mechanisms influenced by protein domain composition, and contribute to unveil new biomedical knowledge about the codifying genes.

GFINDer: genetic disease and phenotype location statistical analysis and mining of dynamically annotated gene lists.

Phenotype analysis is commonly recognized to be of great importance for gaining insight into genetic interaction underlying inherited diseases. However, few computational contributions have been proposed for this purpose, mainly owing to lack of controlled clinical information easily accessible and structured for computational genome-wise analyses. We developed and made available through GFINDer web server an original approach for the analysis of genetic disorder related genes by exploiting the information on genetic diseases and their clinical phenotypes present in textual form within the Online Mendelian Inheritance in Man (OMIM) database. Because several synonyms for the same name and different names for overlapping concepts are often used in OMIM, we first normalized phenotype location descriptions reducing them to a list of unique controlled terms representing phenotype location categories. Then, we hierarchically structured them and the correspondent genetic diseases according to their topology and granularity of description, respectively. Thus, in GFINDer we could implement specific Genetic Disorders modules for the analysis of these structured data. Such modules allow to automatically annotate user-classified gene lists with updated disease and clinical information, classify them according to the genetic syndrome and the phenotypic location categories, and statistically identify the most relevant categories in each gene class. GFINDer is available for non-profit use at http://www.bioinformatics.polimi.it/GFINDer/.

The ALFA4Hearing Model (At-a-Glance Labeling for Features of Apps for Hearing Health Care) to Characterize Mobile Apps for Hearing Health Care.

PURPOSE: The aim of the study was to analyze, by using the ALFA4Hearing model (At-a-Glance Labeling for Features of Apps for Hearing Health Care), a sample of apps over a wide range of services in the hearing health care (HHC) domain in order to take a first picture of the current scenario of apps for HHC. METHOD: We tested 120 apps, and we characterized them by using the ALFA4Hearing model, which includes 29 features in 5 components (Promoters, Services, Implementation, Users, and Descriptive Information). We analyzed (a) the distribution of the 29 features in the sample, (b) the relationship between the Implementation features and the Services provided by the apps, and (c) the distribution of the 29 features in apps for professional use. RESULTS: The analysis of our sample of apps by means of the ALFA4Hearing model highlighted interesting trends and emerging challenges. Also, results suggested many potential opportunities for research and clinical practice, such as greater involvement of stakeholders, improved evidence base, higher technical quality, and usability. CONCLUSIONS: The ALFA4Hearing model is able to represent, at a glance, a large amount of information about apps for HHC, highlighting trends and challenges. It might be useful to HHC professionals as a basis for app characterization and informed decision making.

GFINDer: Genome Function INtegrated Discoverer through dynamic annotation, statistical analysis, and mining.

Statistical and clustering analyses of gene expression results from high-density microarray experiments produce lists of hundreds of genes regulated differentially, or with particular expression profiles, in the conditions under study. Independent of the microarray platforms and analysis methods used, these lists must be biologically interpreted to gain a better knowledge of the patho-physiological phenomena involved. To this end, numerous biological annotations are available within heterogeneous and widely distributed databases. Although several tools have been developed for annotating lists of genes, most of them do not give methods for evaluating the relevance of the annotations provided, or for estimating the functional bias introduced by the gene set on the array used to identify the gene list considered. We developed Genome Functional INtegrated Discoverer (GFINDer), a web server able to automatically provide large-scale lists of user-classified genes with functional profiles biologically characterizing the different gene classes in the list. GFINDer automatically retrieves annotations of several functional categories from different sources, identifies the categories enriched in each class of a user-classified gene list and calculates statistical significance values for each category. Moreover, GFINDer enables the functional classification of genes according to mined functional categories and the statistical analysis is of the classifications obtained, aiding better interpretation of microarray experiment results. GFINDer is available online at http://www.medinfopoli.polimi.it/GFINDer/.

Using Gene Ontology and genomic controlled vocabularies to analyze high-throughput gene lists: three tool comparison.

In genomic and molecular biology domains, controlled vocabularies and ontologies are becoming of paramount importance to integrate and correlate the massive amount of information increasingly accumulating in heterogeneous and distributed databanks. Although at present they are still few and present some issues, they can effectively be used also to biologically annotate genes on a genomic scale and across different species, and to evaluate the relevance of such annotations. Here, we compare three tools using the Gene Ontology and genomic controlled vocabularies to statistically highlight significant biological characteristics of gene sets to help in the biological interpretation of high-throughput experiment results and knowledge discovery from data.

The unexpected high practical value of medical ontologies.

Ontology is no longer a mere research topic, but its relevance has been recognized in several practical fields. Current applications areas include natural language translation, e-commerce, geographic information systems, legal information systems and biology and medicine. It is the backbone of solid and effective applications in health care and can help to build more powerful and more interoperable medical information systems. The design and implementation of ontologies in medicine is mainly focused on the re-organization of medical terminologies. This is obviously a difficult task and requires a deep analysis of the structure and the concepts of such terminologies, in order to define domain ontologies able to provide both flexibility and consistency to medical information systems. The aim of this special issue of Computers in Biology and Medicine is to report the current evolution of research in biomedical ontologies, presenting both papers devoted to methodological issues and works with a more applicative emphasis.

He@lthCo-op: a web-based system to support distributed healthcare co-operative work.

Healthcare is characterized by close collaboration and information sharing among many distinct actors, who co-operate for the patient care in different temporal moments, also at a distance. In this context, availability to care givers of all relevant patient health data and of specific healthcare co-operative work supporting tools is fundamental for best patient treatment. We designed and implemented He@lthCo-op, a web-based modular system supporting co-operative work and patient information secure sharing among healthcare personnel also from remotely located sites. He@lthCo-op enables easily gathering, storing, and accessing patient clinical and personal data anytime and from anywhere an Internet connection is available.

Attempting to Use MetaMap in Clinical Practice: A Feasibility Study on the Identification of Medical Concepts from Italian Clinical Notes.

This study assessed the feasibility of using MetaMap to identify medical concepts from clinical notes written in Italian. We performed two experiments: in "EXP 1", we used MetaMap to annotate Italian texts using a knowledge source consisting of Italian UMLS sources only; in "EXP 2", we used MetaMap to analyze an English unsupervised translated version of the original Italian texts. We considered medical concepts related to three semantic categories: "Disorders", "Findings" and "Symptoms". Average recall, precision and F-measure were equal to 0.53, 0.98 and 0.69 in "EXP 1", and to 0.75, 0.95 and 0.83 in "EXP 2". For both "EXP 1" and "EXP 2" MetaMap showed better performances for the "Disorders" than for "Findings" and "Symptoms". In conclusion, when using MetaMap with the English translation of the Italian clinical notes, we obtained performances good enough to allow using MetaMap in clinical practice. Further investigation about the types of MetaMap's failures could be useful to understand how to improve performances even better.

Advising patients on selecting trustful apps for diabetes self-care.

BACKGROUND: There has been a dramatic increase in mobile apps for diabetes self-care. However, their quality is not guaranteed and patients do not have the appropriate tools for careful evaluation. OBJECTIVE: This work aims to propose a tool to help patients with diabetes select an appropriate app for self-care. METHODS: After identifying the conceptual framework of diabetes self-care, we searched Apple US app store and reviewed diabetes self-care apps, considering both generic and diabetes-specific features. Based on an existing tool for representing the benefits and weaknesses of medical apps, we created the pictorial identification schema/Diabetes Self-care tool, which specifically identified medical apps in the diabetes domain. RESULTS: Of the 952 apps retrieved, 67 were for diabetes self-care, while 26 were excluded because they were not updated in the last 12 months. Of the remaining 41, none cost more than 15 USD, and 36 implemented manual data entry. Basic features (data logging, data representation, and data delivery) were implemented in almost all apps, whereas advanced features (e.g., insulin calculator) were implemented in a small percentage of apps. The pictorial identification schema for diabetes was completed by one patient and one software developer for 13 apps. Both users highlighted weaknesses related to the functionalities offered and to their interface, but the patient focused on usability, whereas the software developer focused on technical implementation. CONCLUSIONS: The Pictorial Identification Schema/Diabetes Self-care is a promising graphical tool for perceiving the weaknesses and benefits of a diabetes self-care app that includes multiple user profile perspectives.

A Design Methodology for Medical Processes.

BACKGROUND: Healthcare processes, especially those belonging to the clinical domain, are acknowledged as complex and characterized by the dynamic nature of the diagnosis, the variability of the decisions made by experts driven by their experiences, the local constraints, the patient's needs, the uncertainty of the patient's response, and the indeterminacy of patient's compliance to treatment. Also, the multiple actors involved in patient's care need clear and transparent communication to ensure care coordination. OBJECTIVES: In this paper, we propose a methodology to model healthcare processes in order to break out complexity and provide transparency. METHODS: The model is grounded on a set of requirements that make the healthcare domain unique with respect to other knowledge domains. The modeling methodology is based on three main phases: the study of the environmental context, the conceptual modeling, and the logical modeling. RESULTS: The proposed methodology was validated by applying it to the case study of the rehabilitation process of stroke patients in the specific setting of a specialized rehabilitation center. The resulting model was used to define the specifications of a software artifact for the digital administration and collection of assessment tests that was also implemented. CONCLUSIONS: Despite being only an example, our case study showed the ability of process modeling to answer the actual needs in healthcare practices. Independently from the medical domain in which the modeling effort is done, the proposed methodology is useful to create high-quality models, and to detect and take into account relevant and tricky situations that can occur during process execution.

Domains of Health IT and Tailoring of Evaluation: Practicing Process Modeling for Multi-Stakeholder Benefits.

This contribution focuses on the heterogeneity and complexity of health information technology services and systems in a multi-stakeholder environment. We propose the perspective of process modeling as a method to break out complexity, represent heterogeneity, and provide tailored evaluation and optimization of health IT systems and services. Two case studies are presented to show how process modeling is needed to fully understand the information flow, thus identifying requirements and specifications for information system re-engineering and interoperability; detect process weaknesses thus designing corrective measures; define metrics as a mean to evaluate and ensure system quality; and optimize the use of resources.

MicroGen: a MIAME compliant web system for microarray experiment information and workflow management.

BACKGROUND: Improvements of bio-nano-technologies and biomolecular techniques have led to increasing production of high-throughput experimental data. Spotted cDNA microarray is one of the most diffuse technologies, used in single research laboratories and in biotechnology service facilities. Although they are routinely performed, spotted microarray experiments are complex procedures entailing several experimental steps and actors with different technical skills and roles. During an experiment, involved actors, who can also be located in a distance, need to access and share specific experiment information according to their roles. Furthermore, complete information describing all experimental steps must be orderly collected to allow subsequent correct interpretation of experimental results. RESULTS: We developed MicroGen, a web system for managing information and workflow in the production pipeline of spotted microarray experiments. It is constituted of a core multi-database system able to store all data completely characterizing different spotted microarray experiments according to the Minimum Information About Microarray Experiments (MIAME) standard, and of an intuitive and user-friendly web interface able to support the collaborative work required among multidisciplinary actors and roles involved in spotted microarray experiment production. MicroGen supports six types of user roles: the researcher who designs and requests the experiment, the spotting operator, the hybridisation operator, the image processing operator, the system administrator, and the generic public user who can access the unrestricted part of the system to get information about MicroGen services. CONCLUSION: MicroGen represents a MIAME compliant information system that enables managing workflow and supporting collaborative work in spotted microarray experiment production.

Inherited disorder phenotypes: controlled annotation and statistical analysis for knowledge mining from gene lists.

BACKGROUND: Analysis of inherited diseases and their associated phenotypes is of great importance to gain knowledge of underlying genetic interactions and could ultimately give clinically useful insights into disease processes, including complex diseases influenced by multiple genetic loci. Nevertheless, to date few computational contributions have been proposed for this purpose, mainly due to lack of controlled clinical information easily accessible and structured for computational genome-wise analyses. To allow performing phenotype analyses of inherited disorder related genes we implemented new original modules within GFINDer http://www.bioinformatics.polimi.it/GFINDer/, a Web system we previously developed that dynamically aggregates functional annotations of user uploaded gene lists and allows performing their statistical analysis and mining. RESULTS: New GFINDer modules allow annotating large numbers of user classified biomolecular sequence identifiers with morbidity and clinical information, classifying them according to genetic disease phenotypes and their locations of occurrence, and statistically analyzing the obtained classifications. To achieve this we exploited, normalized and structured the information present in textual form in the Clinical Synopsis sections of the Online Mendelian Inheritance in Man (OMIM) databank. Such valuable information delineates numerous signs and symptoms accompanying many genetic diseases and it is divided into phenotype location categories, either by organ system or type of finding. CONCLUSION: Supporting phenotype analyses of inherited diseases and biomolecular functional evaluations, GFINDer facilitates a genomic approach to the understanding of fundamental biological processes and complex cellular mechanisms underlying patho-physiological phenotypes.

Apps for hearing healthcare.

The hearing healthcare scenario is rapidly evolving due to the pervasive use of m-Health solutions, in particular mobile apps. This brings along significant advantages and opportunities (e.g., accessibility, affordability, personalized healthcare, patient empowerment) as well as significant potential risks and threats (e.g., safety, misuse, quality issues, privacy). Our research aims at the identification and assessment of apps in the hearing healthcare domain. In this article we present an overview of the current availability, variety, and penetration of hearing-related apps.