【814】Static hotspot analysis and emerging hotspot analysis based on the R library of sfdep

Ref: Emerging Hot Spot Analysis

Ref:

Ref:

Static hotspot analysis

library(tidyverse)
library(sf)
library(openxlsx)
library(ggplot2)
library(tmap)
tmap_mode("view")
library(sfhotspot)

# Set default work directory
setwd("/Users/libingnan/Documents/09-Samsung/12-Polygon-based emerging hotspot analysis")

hist <- read.xlsx("epiwatch_monkeypox.xlsx") %>%
  mutate(PubDate   = as.Date(`insert-timestamp`, origin = "1899-12-30")) %>%
  # filter(PubDate >= as.Date("2020-01-01")) %>%
  mutate(Longitude = as.numeric(long),
         Latitude  = as.numeric(lat)) %>%
  drop_na(Longitude, Latitude) %>%
  # filter(diseases == "covid19") %>%
  # line below removes diseases with less than 5 occurrences
  # group_by(diseases) %>% filter(n() >=5) %>% ungroup() %>%
  #drop_na(diseases) %>%
  st_as_sf(coords = c("Longitude", "Latitude"), crs = 4326) %>%
  st_transform(3857)

results <- hotspot_classify(data = hist,
                            time = PubDate,
                            period = "1 week",
                            cell_size = 500000, # 500 km
                            #cell_size = 200000, # 200 km
                            quiet = F,
                            params = hotspot_classify_params(
                              nb_dist = 0.1) # default is to use points outside of cell.
                            # changed to minimum distance to reduce confusion
)


#autoplot(results)

tm_shape(results, name = "Hotspot Detection") +
  tm_polygons("hotspot_category", title = "Hotspot Category",
              palette = c("persistent hotspot" = "red",
                          "emerging hotspot" = "orange",
                          "intermittent hotspot" = "yellow",
                          "former hotspot" = "darkgreen",
                          "no pattern" = NA),
              alpha = 0.7, lwd = 0.8) +
  
  tm_shape(hist, name = "Reports") +
  tm_dots(jitter=0.1)

结果显示：

Emerging hotspot analysis

library(tidyverse)
library(sf)
library(openxlsx)
library(ggplot2)
library(tmap)
tmap_mode("view")
library(sfhotspot)
library(sfdep)
library(dplyr)

# get directories of files
df_fp <- system.file("extdata", "bos-ecometric.csv", package = "sfdep")
geo_fp <- system.file("extdata", "bos-ecometric.geojson", package = "sfdep")

# read in data
df <- readr::read_csv(df_fp, col_types = "ccidD")

geo <- sf::read_sf(geo_fp)

# Create spacetime object called `bos`
bos <- spacetime(df, geo,
                 .loc_col = ".region_id",
                 .time_col = "time_period")

# conduct EHSA
ehsa <- emerging_hotspot_analysis(
  x = bos,
  .var = "value",
  k = 1,
  nsim = 9
)

# should put geo in the first place, otherwise it will triger the projection error
geo_ehsa <- merge(geo, ehsa, by.x=".region_id", by.y="location")

# tm_shape: Specify the shape object
# tm_polygons: Draw polygons
# "clssification" is a column of hotspot_results

tm_shape(geo_ehsa, name = "Hotspot Detection") +
  tm_polygons("classification", title = "Hotspot Category",
              palette = c("no pattern detected" = "#4576b5"),
              alpha = 0.7, lwd = 0.8)

结果显示：