changes to camare-ready submission

.gitignore

@@ -130,4 +130,13 @@ dmypy.json
 
 # notebooks
 .ipynb
-scripts/2021
+
+# my ignores
+scripts/2021/
+scripts/notebooks/
+ia-env/
+*.bin
+*.pkl
+*.h5
+*.json
+.idea/

2021/README.md

@@ -2,4 +2,4 @@
 
 This folder contains the training, development, and test data of the eHealth-KD 2021 challenge.
 Data will be released in due time.
-Please check [the website](https://ehealthkd.github.io/2021) for more details.
+Please check [the website](https://ehealthkd.github.io/2021) for more details.

2021/eval/testing/README.md

@@ -1,3 +1,11 @@
 # Testing data
 
-This folder will contain the testing data in evaluation format, i.e., divided by scenarios.
+This folder contains the testing data in evaluation format, i.e., divided by scenarios.
+
+The folder `scenario1-main` contains `3000` sentences. The first `2700` are in Spanish and the last `300` are in English. Of these, only `50` in each language will be used for evaluation (the rest are provided to discourage manual annotation), but you must output annotations for all of the sentences, since the reference sentences are shuffled. 
+
+Teams that prefer to participate only in one language can ignore the remaining sentences and not output annotations for them. **However**, make sure to respect character positions to maintain the alignment with the reference annotations. For example, if you are ignoring the Spanish sentences, then your first entity annotation should start around char 195,700 which is roughly where English sentences begin.
+
+Folders `scenario2-taskA` and `scenario3-taskB` contains `50` Spanish sentences and `50` English sentences in that order.
+
+In respect of the ethics of the competition, we kindly ask participants **not** to manually review the testing output, beyond the minimum necessary to guarantee there are no implementation errors. Especially, please do not manually annotate any of the test-set sentences, evaluate your predicted annotations, or make any design decision based on perceived performance of the test set. Doing so would incur in overfitting the testing data and diminish the value of the challenge for all participants.

2021/eval/testing/scenario2-taskA/input.txt

@@ -0,0 +1,100 @@
+Una fractura es una ruptura de un hueso.
+Algunas de las enfermedades transmitidas por picaduras de garrapata son la enfermedad de Lyme, ehrlichiosis, fiebre de las montañas rocallosas y tularemia.
+Para diagnosticar NF1 y NF2 también se usan pruebas genéticas.
+El paladar hendido y el labio leporino son defectos congénitos.
+El asma se trata con dos tipos de medicamentos: medicinas para el alivio rápido para detener los síntomas y medicinas para el control a largo plazo para la prevención de síntomas.
+Los glóbulos rojos suministran oxígeno desde los pulmones a los tejidos y órganos.
+Estos incluyen ungüentos que adormecen las llagas, ablandan las costras o que las secan.
+Los trastornos del cóccix incluyen traumatismos, dolor, infecciones, quistes y tumores.
+Los adultos con diarrea deben beber agua, jugos de frutas, bebidas deportivas, bebidas gaseosas sin cafeína y caldos salados.
+Si usted tiene síntomas de una úlcera péptica, el médico le hará pruebas de sangre, aliento o heces para ver si contiene H. pylori.
+La neuralgia del trigémino generalmente afecta a personas mayores de 50 años, especialmente a mujeres.
+El cáncer de la vesícula puede ser difícil de tratar porque a menudo no se detecta hasta que ya se ha desarrollado.
+La agorafobia es el temor a los lugares públicos y la claustrofobia es el temor a los espacios cerrados.
+Los problemas más serios pueden incluir meningitis, coágulos sanguíneos, inflamación del sistema digestivo y ceguera.
+Los primeros síntomas de la arteritis de células gigantes se parecen a los de la gripe: fatiga, pérdida de apetito y fiebre.
+Pruebas de detección durante el embarazo pueden diagnosticar espina bífida.
+Debido a que es muy común, la mayoría de las personas se contagian con VPH poco después de ser sexualmente activos por primera vez.
+No todas las personas con baja masa ósea tienen osteoporosis, pero están en alto riesgo de tenerla.
+Las articulaciones son los lugares donde se unen dos huesos, como el codo o la rodilla.
+Otras complicaciones a largo plazo de la diabetes incluyen problemas con la piel, problemas en el aparato digestivo, disfunción sexual y problemas en los dientes y las encías.
+La polimiositis causa debilidad en los músculos y en general afecta a los músculos cercanos al tronco del cuerpo.
+Si usted tiene resistencia a la insulina, se acumula un exceso de azúcar en la sangre, preparando el escenario para la aparición de la enfermedad.
+Un ataque cerebrovascular puede causar daño cerebral permanente.
+Si siente dolor y rigidez en su cuerpo o tiene problemas para moverse es posible que padezca de artritis.
+La trombosis venosa profunda o TVP, es un coágulo sanguíneo que se forma en una vena profunda en el cuerpo.
+Varios eventos locales, nacionales y mundiales de todo tipo (conciertos, eventos sociales, deportivos, religiosos, entre otros) han sido suspendidos o aplazados por culpa de la pandemia.
+Christian Lindmeier, portavoz de la Organización Mundial de la Salud (OMS) indicó que es preferible usar paracetamol en lugar de ibuprofeno para paliar los efectos del COVID-19.
+Ali Mohamed Zaki aisló e identificó el virus en un paciente saudí de 60 años.
+Las pruebas originales desarrolladas por los CDC se enviaron el 5 de febrero.
+Para el 28 de febrero de 2020, los mercados bursátiles de todo el mundo acumulaban los mayores descensos de una semana desde la crisis financiera de 2008.
+Una semana después, un hombre italiano repatriado de regreso a Italia desde la ciudad de Wuhan, China, fue hospitalizado y se confirmó que era el tercer caso de COVID 19 en Italia.
+Hasta el 19 de marzo, después de realizar 56,221 pruebas, ha habido 3,269 casos confirmados de COVID-19 en el Reino Unido.
+El 29 de mayo de 2013, la Organización Mundial de la Salud advertía que el virus MERS-CoV es una amenaza a nivel mundial.
+El 7 de marzo, los funcionarios de los CDC confirmaron que el virus se estaba propagando en la comunidad, es decir, que se transmitía de persona a persona con una fuente desconocida, y que esto probablemente continuaría en los próximos días.
+El ministerio ofreció a los hospitales rusos descripciones y guías sobre el mecanismo de acción del tratamiento y las dosis recomendadas.
+En agosto de 2013 el número total de pacientes diagnosticados en el país era de 76, llegando a 127 en noviembre de 2013.
+El 13 de enero se detectó un caso en Tailandia confirmado por pruebas de laboratorio.
+Pero 125 niños, casi el 6 por ciento, desarrollaron una enfermedad muy grave y un niño de 14 años con infección confirmada por coronavirus murió.
+Europa se considera ahora el epicentro de COVID-19 según la Organización Mundial de la Salud a partir del 13 de marzo de 2020.
+Aunque Alemania se encuentra entre los diez países con el mayor número de casos positivos en el mundo, hasta el 13 de marzo solo se registraron ocho muertes, con una tasa de mortalidad del 0,2%.
+El 12 de marzo, el nivel de riesgo del Reino Unido aumentó de moderado a alto.
+El encierro ha causado pánico en la ciudad de Wuhan, y muchos han expresado su preocupación sobre la capacidad de la ciudad para soportar el brote.
+Once municipios en el norte de Italia han sido identificados como los epicentros de los dos principales grupos italianos y puestos en cuarentena.
+La paciente era una azafata china que probablemente había estado en Wuhan días antes de regresar a Viena.
+Debido a la importancia del turismo en la actividad económica italiana (13% del PIB), se teme que esta pandemia tenga consecuencias potencialmente desastrosas.
+Científicos israelíes esperan tener lista una vacuna oral en 90 días, después de pasar por la fase de pruebas de seguridad.
+Las personas recibirán volantes y cuestionarios para ser informados sobre el COVID-19.
+El 31 de diciembre de 2019, las autoridades sanitarias de la ciudad de Wuhan informaron sobre la aparición de veintisiete personas diagnosticadas de síndrome respiratorio agudo grave de origen desconocido.
+La mayoría de los casos están relacionados con viajes al extranjero, en especial a Italia, y en menor grado a Austria.
+La persona, un hombre de unos 50 años, fue dada de alta del hospital el 1 de febrero en buenas condiciones para aislamiento en el hogar.
+Taken together, we conclude that T cells are decreased and exhausted in patients with COVID-19.
+The increased lung macrophage population was present in severe COVID-19 patients
+Firstly, COVID-19 has a relatively long incubation period, which causes a time delay between real dynamic and the dailyobserved case numbers
+Implications of all the available evidence: COVID-19 is currently exhibiting sustained transmission in China
+The pathological process of severe COVID-19 pneumonia is the inflammation reaction characterized by destruction of deep airway and alveolar
+COVID-19 has currently spread to 85 countries and most national authorities have failed to keep it's rapid spread contained
+We assumed that it might be related to bystander activation induced by hyper inflammation, or a delayed response to the infection in severe COVID-19
+The epidemiological curves of COVID-19 in China showed the progression of illness in the outbreak over time from December 8, 2019 up to February 11, 2020, 11 when there were a total of 72314 confirmed cases as the geo-temporal spread of COVID- 19
+Over half (n=30, 56.6%) of the COVID-19 patients exhibited elevated cardiac marker levels
+In addition to the diagnosis value of Ab test, our study revealed a strong positive correlation between clinical severity and antibody titer since 2-week after illness onset, for the first time in COVID-19 patients
+In this study, the median age of 116 COVID-19-confirmed patients was 54 years (IQR, 38-69; range 20-95 years), of which 67 (57.8%) were male
+None of our cases had severe respiratory illness requiring non-invasive or mechanical ventilation, suggesting it will be rare that patients meeting the definition of a suspected case will require enhanced care, at least while COVID-19 itself remains rare in the UK
+Similar to the 2002/2003 outbreak of SARS, severity of COVID-19 disease is associated with increased age and/or a comorbidity, although severe disease is not limited to these risk groups
+The existing spread of COVID-19 is faster than SARS partially because it has more complicate spread dynamics
+COVID-19 is apparently more contagious than SARS and, without protection, people can be easily infected by close contact
+Interestingly, there were 24 (45%) COVID-19 patients with elevation in the levels of one or more myocardial enzymes and markers of myocardial damage whose features did not meet the strict definition of myocardial injury, with most of them having the common NCP type and experiencing quick recovery; however, 58.33% of these patients exhibited LV dysfunction on echo, and long-term outcomes should be observed via follow-up in a future study
+We demonstrate here that COVID-19 patients have very high levels of serum IL-10 following SARS-CoV-2 infection, while also displaying high levels of the PD-1 and Tim-3 exhaustion markers on their T cells, suggesting that IL-10 might be mechanistically responsible
+By assuming these rates of early epidemic growth, we estimate that 130 cities in China have ≥50% chance of having a COVID-19 case imported from Wuhan in the 3 weeks preceding the quarantine
+In our survey, we found that although 81.1% of the parents were paying attention to the COVID-19 epidemic and the situation of the CKD itself, only 4.32% planned to adjust the medication dosage based on their previous medical knowledge and experience
+As mentioned in previous studies, the severity of lymphocytopenia might indicate the severity of COVID-19, under the assumption of SARS-CoV-2 viral could attack and destroy the lymphocyte targetedly
+In a recent analysis, of the 44,672 confirmed COVID-19 cases in China, only 2.1% were in pediatric patients (≤19 years of age)
+Moverever, the levels of elevated ALT and AST are generally not high on admission in our study, indicating the COVID-19 -related liver injury is not serious at the initial stage of infection
+Regarding to laboratory tests, consistent with most reports, peripheral lymphocytopenia is the most crucial and common feature for COVID-19 infection
+Besides, nearly half of COVID-19 patients may not have fever at admission(6) and the lymphocyte count in patients with heart failure can also be decreased
+For a scenario such as the current COVID-19 outbreak, there is unfortunately significant morbidity and mortality, and it is heterogeneously distributed between age groups
+The effective reproductive number, R, of COVID-19 in Hubei was in a range of 2.1 to 4.2 in scenarios based on different assumptions and data sources
+Different with heart failure, COVID-19 was characterized by an inflammatory pulmonary edema and alveolar damage
+Logistic regression analysis indicated that NCP severity was a risk factor for AMI in COVID-19 patients
+Other significant parameters, age and meaningful lower respiratory symptoms, were also included into the COVID-19 EWS
+A high correlation of blood lymphocytes with the disease progression suggested that lymphocyte deficiency or incapacity is the key cellular pathology of COVID-19
+The results also indicate that the relationship between the severity of COVID-19 in a location and life satisfaction depends on individuals' level of exercise, but in a direction opposite to our expectation
+As for COVID-19, several studies have described clinical manifestations including respiratory symptoms, myalgia and fatigue
+The daily statistics showed that lockdown are effective in reduction of incidence of confirmed cases with COVID-19 after about 11 days in china
+Whatever the case, lymphopenia is an indicator of severe COVID-19
+However, whether the limited number of pediatric COVID-19 cases is due to less susceptibility or milder presentation leading to missed detection remains unknown
+The COVID-19 had been affirmed to have human-to-human transmissibility [3] and has now spread worldwide, which raised high attention not only within China but internationally
+The high risk factors of cardiac injury including tachycardia, TnI elevation and low EAT density in CT scan, have been observed in the severe and critical COVID-19 patients
+We found statistically significant evidence of positive diversifying selection (~7% of S -gene sites) along the branch leading to COVID-19
+In addition, another explanation is that some COVID-19 patients may have a past history of chronic kidney disease (CKD) and CKD patients have a proinflammatory milieu and functional defects in innate and adaptive immune cell populations
+The levels of PCT, a marker suggesting the bacterial infection, were not elevated in the patients with COVID-19, suggesting the viral-mediated pneumonia rather than bacteria
+We conclude that a significant reduction of COVID-19 cases is achievable only through a systematic package consisting of enhanced control measures including self-isolation / self-production , effective quarantine and rapid detection / testing
+The true CFR of COVID-19 in China except Hubei Province 137 should fall between 0.8% and 0.9%
+Greater encouragement from health authorities is needed to assimilate COVID-19-related knowledge among HCWs, including doctors
+Interestingly, the numbers of total T cells, CD4 + T and CD8 + T cells are negatively correlated to levels of TNF-α, IL-6 and IL-10, respectively (Figure 2B) , suggesting these cytokines promote T cells decrease in COVID-19 patients
+Unlike pneumonia caused by other pathogens, COVID-19 has some unique characteristics on the CT scan
+These results corresponded to a significantly increased risk of blood group A for COVID-19 with an OR of 1.279 (95% CI Table 1 )
+In general, COVID-19 in severe cases progresses rapidly to lethal complications (ARDS, septic shock, acute cardiac injury, refractory metabolic acidosis and so on), even leading to death
+According to the CT reports, if a CT scan was COVID-positive, its ground-truth label was 1; otherwise, the label was 0
+We next examined 113 patients with COVID-19 from another hospital in Wuhan City, the Renmin Hospital of Wuhan University, and found a similar risk distribution trend of ABO blood groups for the infection
+For a given COVID-19 case with symptom onset on the t-th day, the number of secondary cases follows a negative binomial (NB) distribution with a mean at Reff(t) and a dispersion parameter at k