ABSTRACT
Confinement during COVID-19 has caused serious effects on agriculture all over the world. As one of the efficient solutions, mechanical harvest/auto-harvest that is based on object detection and robotic harvester becomes an urgent need. Within the auto-harvest system, robust few-shot object detection model is one of the bottlenecks, since the system is required to deal with new vegetable/fruit categories and the collection of large-scale annotated datasets for all the novel categories is expensive. There are many few-shot object detection models that were developed by the community. Yet whether they could be employed directly for real life agricultural applications is still questionable, as there is a context-gap between the commonly used training datasets and the images collected in real life agricultural scenarios. To this end, in this study, we present a novel cucumber dataset and propose two data augmentation strategies that help to bridge the context-gap. Experimental results show that 1) the state-of-the-art few-shot object detection model performs poorly on the novel 'cucumber' category;and 2) the proposed augmentation strategies outperform the commonly used ones.
ABSTRACT
Rationale: Patients with chronic intestinal failure (CIF) have a high probability of long-term survival on HPN. In the general population, patients with COVID-19 can experience digestive symptoms. Taking into account CIF patients characteristics, we hypothesized that a COVID-19 infection can determine delayed diagnosis and care, more severe complications, and atypical clinical presentations in CIF population. The objective of this study was to establish the clinical presentation of COVID 19 infection in CIF patients and their outcome. Methods: A prospective and retrospective national multicenter study, including all CIF patients (HPN > 3 months) with confirmed diagnosis of COVID-19 infection (by lung CTscan or PCR) were included. All approved HPN centre for adults and children participated. Results: The Inclusion period was from the 01/02/ 2020 to 05/05/2020. Among approximately 1000 CIF adult patients and 500 children, 9 (7F/2M) adult patients with a median BMI of 20.8 (±3)kg/m2 were diagnosed with Covid-19 infection, none in pediatric population. The median age was 56 (±23)years. The cause of CIF was short bowel syndrome (7/9) and motility disorder (2/9) with a 5,9 (±6) years of HPN duration. COVID 19 was confirmed by the association of positive COVID-19 PCR and CTscan in 6 cases, a positive PCR performed in a private laboratory in 2 cases and typical pulmonary lesions on CT-scan despite a negative PCR in one case. Clinical symptoms firstly described were fever (5/9), anosmia (2/9), headache (2/9), dyspnea (5/9) and digestive symptoms (4/9). Six patients were hospitalized, and 2 of them required intensive care. Two patients presented severe complications;one hyperosmolar hyperglycemic state in a patient without previous diabetes diagnosis and one death probably due to a massive pulmonary embolism 1 month after COVID-19 infection in a context of pulmonary arterial hypertension. Conclusion: In the context of COVID pandemia, the prevalence of COVID 19 diagnosis in CIF adult population seems to be around 9/1000. Currently, no case was diagnosed in pediatric CIF patients. Since the initial presentation could be unusual in patients without known risk factors, expert centers should be alerted. As the infectious disease continues to spread around the world, a worldwide international would be necessary to deepen those data. Disclosure of Interest: None declared