A dataset of skin lesion images collected in Argentina for the evaluation of AI tools in this population.

In recent years, numerous dermatological image databases have been published to make possible the development and validation of artificial intelligence-based technologies to support healthcare professionals in the diagnosis of skin diseases. However, the generation of these datasets confined to certain countries as well as the lack of demographic information accompanying the images, prevents having a real knowledge of in which populations these models could be used. Consequently, this hinders the translation of the models to the clinical setting. This has led the scientific community to encourage the detailed and transparent reporting of the databases used for artificial intelligence developments, as well as to promote the formation of genuinely international databases that can be representative of the world population. Through this work, we seek to provide details of the processing stages of the first public database of dermoscopy and clinical images created in a hospital in Argentina. The dataset comprises 1,616 images corresponding to 1,246 unique lesions collected from 623 patients.

Radiomics and Machine Learning for prediction of two-year disease-specific mortality and KRAS mutation status in metastatic colorectal cancer.

PURPOSE: Colorectal cancer is usually accompanied by liver metastases. The prediction of patient evolution is essential for the choice of the appropriate therapy. The aim of this study is to develop and evaluate machine learning models to predict KRAS gene mutations and 2-year disease-specific mortality from medical images. METHODS: Clinical and follow-up information was collected from patients with metastatic colorectal cancer who had undergone computed tomography prior to liver resection. The dominant liver lesion was segmented in each scan and radiomic features were extracted from the volumes of interest. The 65% of the cases were employed to perform feature selection and to train machine learning algorithms through cross-validation. The best performing models were assembled and evaluated in the remaining cases of the cohort. RESULTS: For the mortality model development, 101 cases were used as training set (64 alive, 37 deceased) and 35 as test set (22 alive, 13 deceased); while for KRAS mutation models, 55 cases were used for training (31 wild-type, 24 mutated) and 30 for testing (17 wild-type, 13 mutated). The ensemble of top performing models resulted in an area under the receiver operating characteristic curve of 0.878 for mortality and 0.905 for KRAS prediction. CONCLUSIONS: Predicting the prognosis of patients with metastatic colorectal cancer is useful for making timely decisions about the best treatment options. This study presents a noninvasive method based on quantitative analysis of baseline images to identify factors influencing patient outcomes, with the aim of incorporating these tools as support systems.

Efecto de la inactivación de las órdenes múltiples para solicitud de estudios complementarios en una central de emergencias.

Introducción. La plantilla de órdenes múltiples es una herramienta informática que podría producir consecuencias inadvertidas pese a sus innumerables beneficios. Nos propusimos explorar el efecto de su inactivación sobre las solicitudes de estudios complementarios y los costos asociados. Métodos. Corte transversal en la Central de Emergencias de Adultos del Hospital Italiano de Buenos Aires, que incluyó muestra consecutiva de consultas pre-intervención (Enero-Febrero 2020) y post-intervención (2021). Mediante el uso de bases secundarias, las variables incluidas fueron los débitos administrativos y sus respectivos precios de facturación. Resultados. Hubo 27.671 consultas en 2020 con una mediana de valor total de 474$, y 20.819 con 1.639$ en 2021. Tras el análisis restringido al área de consultorios de moderada complejidad (excluyendo consultas por COVID-19), se encontró: una disminución en la mediana del número de prácticas por consulta (mediana de 11 vs 10, p=0,001), una disminución en la solicitud de al menos una práctica de laboratorio (45% versus 39%, p=0,001), sin encontrar cambios significativos en costos globales (mediana 1.419$ vs 1.081$; p=0,122) ni en costos específicos de laboratorio (mediana 1.071$ vs 1.089$, p=0,710). Conclusión. Pese a la inflación interanual, se logró una reducción significativa en el número de prácticas y se mantuvieron los costos globales por consulta. Estos hallazgos demuestran la efectividad de la intervención, pero serán necesarias medidas educativas que apunten al recordatorio de los potenciales daños en la sobreutilización, y los costos sanitarios de los estudios innecesarios.

Effect of computerized physician order entry inactivation to order complementary studies in an emergency department / [Efecto de la inactivación de las órdenes múltiples para solicitud de estudios complementarios en una central de emergencias.]

Introduction: The computerized provider order entry (CPOE) is a computing tool that could lead to unintended consequences despite its myriad benefits. We aimed to explore the effect of its inactivation on requests for complementary studies and the associated costs. Methods: Cross sectional study at the Emergency Department of Hospital Italiano de Buenos Aires, which included a consecutive sample of pre-intervention (January-February 2020) and post-intervention (2021) consultations. Using secondary bases, the variables included were administrative debits and their respective billing prices. Results: There were 27,671 consultations in 2020 with a total median value of $474, and 20,819 with $1,639 in 2021. After the analysis restricted to the area of ​​moderately complex clinics (excluding COVID-19 consultations), the following was found: a decrease in the median number of practices per consultation (median of 11 vs. 10, p=0.001), a decrease in the request for at least one laboratory practice (45% vs. 39%, p=0.001), without finding significant changes in global costs (median $1,419 vs. $1,081; p=0.122) or in specific laboratory costs (median $1,071 vs. $1,089, p=0.710). Conclusion: Despite inflation, a significant reduction in the number of practices was achieved and overall costs per consultation were maintained. These findings demonstrate the effectiveness of the intervention, but an educational intervention aimed at reminding the potential harm of overuse and the health costs of unnecessary studies will be necessary.

Introducción: La plantilla de órdenes múltiples es una herramienta informática que podría producir consecuencias inadvertidas pese a sus innumerables beneficios. Nos propusimos explorar el efecto de su inactivación sobre las solicitudes de estudios complementarios y los costos asociados. Métodos: Corte transversal en la Central de Emergencias de Adultos del Hospital Italiano de Buenos Aires, que incluyó muestra consecutiva de consultas pre-intervención (Enero-Febrero 2020) y post-intervención (2021). Mediante el uso de bases secundarias, las variables incluidas fueron los débitos administrativos y sus respectivos precios de facturación. Resultados: Hubo 27.671 consultas en 2020 con una mediana de valor total de 474$, y 20.819 con 1.639$ en 2021. Tras el análisis restringido al área de consultorios de moderada complejidad (excluyendo consultas por COVID-19), se encontró: una disminución en la mediana del número de prácticas por consulta (mediana de 11 vs 10, p=0,001), una disminución en la solicitud de al menos una práctica de laboratorio (45% versus 39%, p=0,001), sin encontrar cambios significativos en costos globales (mediana 1.419$ vs 1.081$; p=0,122) ni en costos específicos de laboratorio (mediana 1.071$ vs 1.089$, p=0,710). Conclusión: Pese a la inflación interanual, se logró una reducción significativa en el número de prácticas y se mantuvieron los costos globales por consulta. Estos hallazgos demuestran la efectividad de la intervención, pero serán necesarias medidas educativas que apunten al recordatorio de los potenciales daños en la sobreutilización, y los costos sanitarios de los estudios innecesarios.

Can CPOE Based on Electronic Order Sets Cause Unintended Consequences (Expensive and Unnecessary Tests) at the Emergency Department?

Computerized Provider Order Entry (CPOE) systems may cause unintended consequences. This study aimed to describe the on-going system for CPOE order sets, and to explore an economic evaluation at the Emergency Department. First, we developed a costs dashboard which showed us the significant and excessive use of medical tests per consultation. We identified the top 10 most widely used and most expensive tests. Additionally we noticed that the labs seemed to continually increase. Then, we found that 27% of the consultations have at least one item of laboratory practice between January and February 2020, and this represents more than 80% of the consultation costs. Health care spending has reached epic proportions globally. We think that it is time to rethink effective strategies. Maybe it is time to deactivate/remove electronic order sets (EOSs) and the functionality to develop and create their own "private" order sets, in order to eliminate waste and inefficiencies.

Using Informatics Framework in Immunosuppressed or Transplanted Patients for a Preferential Care Route at the Emergency Department.

For immunosuppressed or transplanted patients, appropriate triage is a timely topic, especially in the Emergency Department (ED) of a high-volume referral center. We implemented a new Program called Rapid Clinical Care by Internal Medicine Specialists, as a preferential care route for these patients, which combines the proposed informatics framework in the field of total quality management in the healthcare units, as an example of digital technologies that can improve processes in the clinical routine. Our study aimed to describe waiting-time and attention-time in ED and to explore the effect on patients' clinical outcomes after discharge. Findings were: shortened waiting time (median of 8 minutes versus 21, p<0.001), improved ED on-call time (median of 2 hours compared to 4, p<0.001), and greater follow-up after discharge, measured as 1-week scheduled-visits rate (69% with 95%CI 63-75; compared to 43% with 95%CI 35-51; p<0.001).

Chest x-ray automated triage: A semiologic approach designed for clinical implementation, exploiting different types of labels through a combination of four Deep Learning architectures.

BACKGROUND AND OBJECTIVES: The multiple chest x-ray datasets released in the last years have ground-truth labels intended for different computer vision tasks, suggesting that performance in automated chest x-ray interpretation might improve by using a method that can exploit diverse types of annotations. This work presents a Deep Learning method based on the late fusion of different convolutional architectures, that allows training with heterogeneous data with a simple implementation, and evaluates its performance on independent test data. We focused on obtaining a clinically useful tool that could be successfully integrated into a hospital workflow. MATERIALS AND METHODS: Based on expert opinion, we selected four target chest x-ray findings, namely lung opacities, fractures, pneumothorax and pleural effusion. For each finding we defined the most suitable type of ground-truth label, and built four training datasets combining images from public chest x-ray datasets and our institutional archive. We trained four different Deep Learning architectures and combined their outputs with a late fusion strategy, obtaining a unified tool. The performance was measured on two test datasets: an external openly-available dataset, and a retrospective institutional dataset, to estimate performance on the local population. RESULTS: The external and local test sets had 4376 and 1064 images, respectively, for which the model showed an area under the Receiver Operating Characteristics curve of 0.75 (95%CI: 0.74-0.76) and 0.87 (95%CI: 0.86-0.89) in the detection of abnormal chest x-rays. For the local population, a sensitivity of 86% (95%CI: 84-90), and a specificity of 88% (95%CI: 86-90) were obtained, with no significant differences between demographic subgroups. We present examples of heatmaps to show the accomplished level of interpretability, examining true and false positives. CONCLUSION: This study presents a new approach for exploiting heterogeneous labels from different chest x-ray datasets, by choosing Deep Learning architectures according to the radiological characteristics of each pathological finding. We estimated the tool's performance on the local population, obtaining results comparable to state-of-the-art metrics. We believe this approach is closer to the actual reading process of chest x-rays by professionals, and therefore more likely to be successful in a real clinical setting.

Asynchronous Dermatology Teleconsultations Using a Personal Health Record.

Health information and communication technologies such as telemedicine provide alternatives for patient and physician communication. An increasing number of patients, providers and institutions are using this technologies to seek or provide health care. Asynchronous consultations requires a service of storing and forwarding health related information by the patient to the specialist physician or other healthcare provider. Dermatology is one of the major medical specialties in which telemedicine as proven benefits. There are many described deployments of asynchronous teleconsultation for dermatology, but in most cases telemedicine applications work as stand alone solutions. The present work describes the design and deployment of an asynchronous dermatological teleconsultation service which uses the interaction of the Electronic Health Record and the Personal Health Record in a high complexity university hospital.

Validating the Access to an Electronic Health Record: Classification and Content Analysis of Access Logs.

Electronic Health Records (EHRs) have made patient information widely available, allowing health professionals to provide better care. However, information confidentiality is an issue that continually needs to be taken into account. The objective of this study is to describe the implementation of rule-based access permissions to an EHR system. The rules that were implemented were based on a qualitative study. Every time users did not meet the specified requirements, they had to justify access through a pop up window with predetermined options, including a free text option ("other justification"). A secondary analysis of a deidentified database was performed. From a total of 20,540,708 hits on the electronic medical record database, 85% of accesses to the EHR system did not require justification. Content analysis of the "Other Justification" option allowed the identification of new types of access. At the time to justify, however, users may choose the faster or less clicks option to access to EHR, associating the justification of access to the EHR as a barrier.

[Changes in physicians' attitudes to computerized ambulatory medical record systems: a longitudinal qualitative study]. / Modificaciones en las perspectivas de los médicos sobre el registro médico electrónico: investigación cualitativa longitudinal.

OBJECTIVES: To explore physicians' beliefs about a computerized ambulatory medical record system at different stages of its implementation. METHODS: We performed a longitudinal qualitative in-depth interview study (July 2001 to December 2003) in the Hospital Italiano, Buenos Aires, Argentina. Semi-structured interviews were conducted in 20 primary care cardiologists purposively selected before, during and after the system's implementation process (10 interviews per stage). The interviews were independently analyzed by 2 researchers, who jointly designed an agreed category list. RESULTS: Both before and during the first stage of the implementation process, the physicians expected that that the system would improve healthcare-related administration and increase accessibility to individual data. However, they did not foresee that the system's shared information could modify the clinical aspects of patient care. By the end of the implementation process, the physicians realized that the system provided them with a broader perspective on their patients, which in turn improved their own professional performance. Throughout the implementation, the physicians were against using the computer while the patient was present. This opposition prevented them from regarding the system as part of the medical consultation and from considering data from the system as direct patient-related signs. CONCLUSIONS: The system's implementation modified the physicians' views on computerized ambulatory medical records, as they eventually considered them as an ancillary tool to clinical activity. The value assigned to the system depends on its relevance within the institutional framework.