• Aucun résultat trouvé

Spatial aggregation and cluster detection

N/A
N/A
Protected

Academic year: 2021

Partager "Spatial aggregation and cluster detection"

Copied!
15
0
0

Texte intégral

(1)

Spatial aggregation and cluster detection

Facundo Muñoz

facundo.munoz@cirad.fr

 famuvie

Biostatistics for epidemiology

December 2018

(2)

John Snow's map on the 1854 London

Cholera Outbreak

(3)

Water pumps and deaths

(4)
(5)

Key questions

Do cases tend to occur in

clusters

?

(assessment of aggregation)

Is there an

unusual group

of cases?

(cluster detection)

Is the case occurrence

related

with something else?

(risk factor identification)

(6)

Key questions

Do cases tend to occur in

clusters

?

(assessment of aggregation)

Is there an

unusual group

of cases?

(cluster detection)

Is the case occurrence

related

with something else?

(risk factor identification)

(7)

in the previous terms,

What do you think Snow did?

(8)

Snow.deaths %>%

ggplot(aes(x, y)) + geom_point() +

coord_fixed() +

labs(x = NULL, y = NULL)

Visualisation

(9)

Snow.deaths %>% ggplot(aes(x, y)) + stat_density_2d( aes(fill = stat(level)), geom = "polygon" ) + geom_point(alpha = .5) + scale_fill_viridis_c( name = "Density" ) + coord_fixed() +

labs(x = NULL, y = NULL)

Point density estimation

(10)

Snow.deaths %>% ggplot(aes(x, y)) + stat_density_2d( aes(fill = stat(level)), geom = "polygon" ) + # geom_point(alpha = .5) + geom_point( data = Snow.pumps, shape = 13, size = 2, col = "blue" ) + scale_fill_viridis_c( name = "Density"

(11)

Risk factor assessment

Note: despite the significant effect, it does not prove causality (right?)

(12)

Objectives for the session

Discuss methods and tools for quantification and detection of global

(13)

Objectives for the session

Discuss methods and tools for quantification and detection of global

aggregation, local clusters and risk factor evaluation

Both in spatial and spatio-temporal settings

(14)

Objectives for the session

Discuss methods and tools for quantification and detection of global

aggregation, local clusters and risk factor evaluation

Both in spatial and spatio-temporal settings

(15)

Objectives for the session

Discuss methods and tools for quantification and detection of global

aggregation, local clusters and risk factor evaluation

Both in spatial and spatio-temporal settings

For both areal and point data

Operational approach

: work out 3 case-studies and discussing topics as needed

Références

Documents relatifs

Estimez et ensuite mesurez `a l’aide d’un rapporteur la valeur de chaque

Face detection based on generic local descriptors and spatial constraints.. Veronika Vogelhuber,

In [19], state-of-the-art deep learning methods for polyp detection, localization, and segmentation were compared on the basis of both accuracy and speed with extensive experiments

Keywords: Cluster detection test, Type I error, Simulation study, Edge effect, Spatial

In this section, we present the evaluation study that we carried out to compare the performance of the previously described aggregation schemes in terms of the accuracy of the

In recent work we have simulated irreversible diffusion limited cluster-cluster aggregation (DLCA) from the ini- tial state of monomers to the final state where all particles form

ةماعلا ةتمالخا 168 ينب ةرشابلما ةيطلخا ةقلاعلا عبأ ةدوج دا ،ةيسومللما( فطاعتلا و ةباجتسلاا ،ديكأتلا ،ةيدامتعلاا ) عم ،ةكردلما ةدولجا و ةقلاع ىلع يرغ ةرشابم

In the case where the user would provide a data table and the table of the transformation parameter set, the transformed variables are obtained using the trans function.. A