Thrombolysis hospital vs. benchmark diagrams and patient vignettes#

In this notebook we present means of comparing use of thrombolysis in a particular hospital to our benchmark set of hospitals. At the hospital level we illustrate the comparison with Graph and Venn diagrams. We then pick individual patients which illustrate something about that hospital, and embellish the SSNAP data with synthetic details in order to provide a discussion vignette for physicians.

Here we illustrate, though hospital-level diagrams and patient-level vignettes, two example areas of differences in treatment decisions:

A hospital with lower-than-usual use of thrombolysis in patients with mild stroke
A hospital with lower-than-usual use of thrombolysis in patients with prior disability

Import libraries#

import os
import pandas as pd
pd.set_option('display.max_rows', 100)
pd.set_option('display.max_columns', 110)
import numpy as np
import graphviz as gv

import matplotlib.pyplot as plt
import matplotlib.ticker as ticker
from matplotlib_venn import venn2, venn2_circles
import matplotlib.gridspec as gridspec

from IPython.display import Image 

Function for Venn diagram plotting#

Venn diagrams are plotted using the matplotlib-venn library (pip install matplotlib-venn).

def plot_venn(df, hospital, save=False):
    
    patients = df.loc[df['StrokeTeam']==hospital]
    
    throm = len(patients.loc[patients['S2Thrombolysis']==1])
    majority_throm = len(patients.loc[patients['Benchmark']==1])
    
    no_throm = len(patients.loc[patients['S2Thrombolysis']==0])
    majority_no_throm = len(patients.loc[patients['Benchmark']==0])
    
    agree_throm = len(patients.loc[(patients['S2Thrombolysis'] == 1) & (patients['Benchmark']==1)])
    agree_no_throm = len(patients.loc[(patients['S2Thrombolysis'] == 0) & (patients['Benchmark']==0)])
    
    
    
    rate = round(throm*100 / (throm + no_throm),1)
    agree = round((agree_throm + agree_no_throm)/(throm + no_throm),1)
    
    
    fig, ax = plt.subplots(1,3, figsize = (15,5))
    
    ax[2].axis('off')
    
    ax[2].text(0.3, 0.8, 'N Patients: {0}'.format(len(patients)), horizontalalignment='left',
        verticalalignment='center', size=18)
    
    ax[2].text(0.3, 0.7, 'Thrombolysis rate (%): {0}'.format(rate), horizontalalignment='left',
        verticalalignment='center', size=18)
    
    ax[2].text(0.3, 0.6, 'Agreement (%): {0}'.format(agree*100), horizontalalignment='left',
        verticalalignment='center', size=18)
    
    
    ax[0].set_title('Thrombolysed', fontsize=18)
    
    f1 = venn2(subsets = (throm-agree_throm, majority_throm-agree_throm, agree_throm), 
          set_labels = ('Hospital', 'Benchmark'), set_colors=('yellow', 'blue'), alpha = 0.7, ax=ax[0])
    f1.get_patch_by_id('11').set_color('green')
    
    venn2_circles(subsets = (throm-agree_throm, majority_throm-agree_throm, agree_throm), 
                  lw=1, alpha=0.5, ax=ax[0])
    
    for text in f1.set_labels:
        text.set_fontsize(16)
        text.set_fontweight('bold')
    for text in f1.subset_labels:
        text.set_fontsize(16)
    
    ax[1].set_title('Not Thrombolysed', fontsize=18)
    
    f2 = venn2(subsets = (no_throm-agree_no_throm, majority_no_throm-agree_no_throm, agree_no_throm),
          set_labels = ('Hospital', 'Benchmark'), set_colors=('yellow', 'blue'), alpha = 0.7, ax=ax[1])
    f2.get_patch_by_id('11').set_color('green')
    
    venn2_circles(subsets = (no_throm-agree_no_throm, majority_no_throm-agree_no_throm, agree_no_throm),
                  lw=1, alpha=0.5, ax=ax[1])
    
    for text in f2.set_labels:
        text.set_fontsize(16)
        text.set_fontweight('bold')
    for text in f2.subset_labels:
        text.set_fontsize(16)
    
    if save==True:
        
        plt.savefig('./figures/vignettes/venn_{0}.jpg'.format(hospital), dpi=300)
        
    plt.show()

Function for decision graph plotting#

Decision graph plots are performed using graphviz (pip install graphviz).

def plot_graph(df, hospital, save=False):
    
    patients = df.loc[df['StrokeTeam']==hospital]
    
    throm = len(patients.loc[patients['S2Thrombolysis']==1])
    
    no_throm = len(patients.loc[patients['S2Thrombolysis']==0])
    
    
    agree_throm = len(patients.loc[(patients['S2Thrombolysis'] == 1) & (patients['Benchmark']==1)])
    agree_no_throm = len(patients.loc[(patients['S2Thrombolysis'] == 0) & (patients['Benchmark']==0)])
    
    disagree_throm = len(patients.loc[(patients['S2Thrombolysis'] == 1) & (patients['Benchmark']==0)])
    disagree_no_throm = len(patients.loc[(patients['S2Thrombolysis'] == 0) & (patients['Benchmark']==1)])
    
    # N prop to w^2/4, h^2/4
    
    dot = gv.Digraph(comment='{0}'.format(hospital))
    
    dot.attr('node', shape='circle', width='2', height='2', fixedsize='true')
    
    dot.node('H', 'Hospital\nAdmissions:\n{0}'.format(len(patients)))

    dot.node('T', 'Hospital\nThrombolysed:\n{0}'.format(throm),\
             style='filled', fillcolor='#FFFD6B')
    
    dot.node('N', 'Hospital Not\nThrombolysed:\n{0}'.format(no_throm),\
            style='filled', fillcolor='#FFFD6B')
    
    dot.node('A', 'Benchmark\nThrombolysed:\n{0}'.format(agree_throm),\
             style='filled', fillcolor='#00CC0F')
    dot.node('B', 'Benchmark Not\nThrombolysed:\n{0}'.format(disagree_throm), \
            style='filled', fillcolor='#4E73F9')
    
    dot.node('C', 'Benchmark Not\nThrombolysed:\n{0}'.format(agree_no_throm),\
             style='filled', fillcolor='#00CC0F')
    dot.node('D', 'Benchmark\nThrombolysed:\n{0}'.format(disagree_no_throm),\
             style='filled', fillcolor='#4E73F9')

    dot.edges(['HT', 'HN', 'TA', 'TB', 'NC', 'ND'])

    
    #dot.render('./figures/vignettes/graph_{0}.gv'.format(hospital), view=True) 
    dot.render('./figures/vignettes/graph_{0}.gv'.format(hospital), view=False, format='png')
    
    return dot

Load data#

# Load dataset
data = pd.read_csv('./data/vignette_data.csv', index_col=0)

data.shape

(88928, 107)

Identify Confident Misclassifications#

Identify patients where model has 90% probability of thrombolysis, but patient was not given thrombolysis.

confident = data.loc[(data['S2Thrombolysis']==0) & (data['Probability']>=0.9)]

confident

	StrokeTeam	S1AgeOnArrival	S1OnsetToArrival_min	S2RankinBeforeStroke	LocQuestions	LocCommands	BestGaze	Visual	FacialPalsy	MotorArmLeft	MotorArmRight	MotorLegLeft	MotorLegRight	Sensory	BestLanguage	Dysarthria	ExtinctionInattention	S2NihssArrival	S2BrainImagingTime_min	MoreEqual80y_No	MoreEqual80y_Yes	S1Gender_Female	S1Gender_Male	S1Ethnicity_Other	S1Ethnicity_White	S1OnsetInHospital_No	S1OnsetDateType_Precise	S1OnsetTimeType_Precise	S1ArriveByAmbulance_Yes	S1AdmissionHour_03:00 to 05:59	S1AdmissionHour_12:00 to 14:59	S1AdmissionHour_18:00 to 20:59	S1AdmissionDay_Friday	S1AdmissionDay_Saturday	S1AdmissionDay_Thursday	S1AdmissionDay_Tuesday	S1AdmissionDay_Wednesday	S1AdmissionQuarter_Q1	S1AdmissionQuarter_Q2	S1AdmissionQuarter_Q3	S1AdmissionQuarter_Q4	S1AdmissionYear_2016	S1AdmissionYear_2017	S1AdmissionYear_2018	CongestiveHeartFailure_No	Hypertension_No	Hypertension_Yes	AtrialFibrillation_No	AtrialFibrillation_Yes	Diabetes_No	Diabetes_Yes	StrokeTIA_No	StrokeTIA_Yes	AFAntiplatelet_No	AFAntiplatelet_missing	AFAnticoagulent_No	AFAnticoagulent_No but	AFAnticoagulent_missing	AFAnticoagulentVitK_No	AFAnticoagulentVitK_missing	AFAnticoagulentDOAC_No	AFAnticoagulentDOAC_missing	AFAnticoagulentHeparin_No	AFAnticoagulentHeparin_missing	S2NewAFDiagnosis_No	S2NewAFDiagnosis_Yes	S2NewAFDiagnosis_missing	S2StrokeType_Infarction	S2TIAInLastMonth_No	S2TIAInLastMonth_No but	S2TIAInLastMonth_missing	Benchmark	Majority_90	Majority_80	Predicted	Probability	AvgProbability
28106	TQQYU0036V	67.5	84.0	1	0.0	0.0	2.0	0.0	2.0	4.0	0.0	4.0	0.0	2.0	0.0	1.0	1.0	16.0	13.0	1	0	1	0	0	1	1	1	1	1	0	1	0	1	0	0	0	0	0	1	0	0	0	0	1	1	1	0	1	0	1	0	1	0	0	1	1	0	0	1	0	1	0	1	0	0	1	0	1	0	0	1	1	1	1	1	0.91	0.71
29126	PAOUI8935Z	62.5	152.0	0	0.0	0.0	0.0	0.0	1.0	3.0	0.0	3.0	0.0	1.0	1.0	1.0	0.0	10.0	1.0	1	0	0	1	0	1	1	1	1	1	1	0	0	0	0	1	0	0	0	0	0	1	0	1	0	1	1	0	1	0	1	0	0	1	0	1	0	0	1	0	1	0	1	0	1	0	0	1	1	0	0	1	1	1	1	1	0.92	0.58
46312	MHMYL4920B	72.5	80.0	0	2.0	0.0	1.0	1.0	2.0	3.0	0.0	3.0	0.0	1.0	1.0	1.0	0.0	15.0	5.0	1	0	0	1	0	1	1	1	1	1	0	0	1	1	0	0	0	0	1	0	0	0	1	0	0	1	1	0	1	0	1	0	0	1	0	1	0	0	1	0	1	0	1	0	1	0	0	1	1	1	0	0	1	1	1	1	0.91	0.65
46870	MHMYL4920B	77.5	50.0	0	0.0	0.0	0.0	0.0	1.0	1.0	0.0	1.0	0.0	0.0	0.0	1.0	0.0	4.0	10.0	1	0	1	0	0	1	1	1	1	1	0	1	0	0	0	0	0	1	0	1	0	0	0	0	1	1	1	0	1	0	1	0	1	0	0	1	1	0	0	1	0	1	0	1	0	1	0	0	1	0	0	1	1	0	0	1	0.94	0.53
48408	ITFRR7678Y	77.5	86.0	0	0.0	0.0	0.0	0.0	2.0	0.0	2.0	0.0	1.0	0.0	1.0	2.0	0.0	8.0	11.0	1	0	0	1	0	1	1	1	1	1	0	0	1	0	1	0	0	0	0	0	1	0	1	0	0	1	0	1	1	0	0	1	1	0	0	1	0	0	1	0	1	0	1	0	1	0	0	1	1	0	0	1	1	1	1	1	0.90	0.66
56912	BQZGT7491V	82.5	104.0	0	1.0	0.0	0.0	0.0	0.0	1.0	0.0	4.0	0.0	1.0	0.0	0.0	0.0	7.0	23.0	0	1	0	1	0	1	1	1	1	1	0	1	0	0	1	0	0	0	0	0	0	1	1	0	0	1	1	0	1	0	0	1	1	0	0	1	0	0	1	0	1	0	1	0	1	0	0	1	1	0	0	1	1	0	0	1	0.90	0.49
57186	BQZGT7491V	77.5	85.0	0	2.0	1.0	0.0	0.0	1.0	0.0	2.0	0.0	2.0	0.0	2.0	2.0	0.0	12.0	23.0	1	0	0	1	0	1	1	1	1	1	0	0	1	0	0	0	0	1	1	0	0	0	0	1	0	1	0	1	1	0	1	0	0	1	0	1	0	0	1	0	1	0	1	0	1	0	0	1	1	0	1	0	1	1	1	1	0.90	0.62
70756	FPTBM7594G	72.5	56.0	0	2.0	0.0	0.0	0.0	2.0	0.0	2.0	0.0	3.0	1.0	3.0	2.0	1.0	16.0	38.0	1	0	1	0	1	0	1	1	1	1	0	1	0	1	0	0	0	0	0	0	0	1	0	1	0	1	1	0	1	0	1	0	1	0	0	1	0	0	1	0	1	0	1	0	1	0	0	1	1	0	0	1	1	1	1	1	0.94	0.66
87345	QQUVD2066Z	92.5	67.0	3	0.0	2.0	2.0	2.0	2.0	1.0	4.0	2.0	3.0	0.0	3.0	2.0	2.0	25.0	10.0	0	1	1	0	0	1	1	1	1	1	0	0	1	0	0	0	1	0	1	0	0	0	0	1	0	1	0	1	0	1	1	0	1	0	1	0	0	1	0	0	1	0	1	0	1	0	0	1	1	0	0	1	1	0	0	1	0.90	0.44

Patient vignette for a confident misclassification

Patient vignettes are intended to provide a centre for discussion with physcicians

Patient at iloc 3 in the confident mislassification table:

Age: 77.5
Onset to arrival: 50 mins
Rankin before: 0
Facial palsy: 1
Motor arm right: 1
Motor leg right: 1
Dysarthia: 1
NIHSS: 4
Brain imaging time: 10 mins
White Female
Precise onset time
Admitted between 12 and 3pm on Wednesday, quarter 2, 2018
No Diabetes, Hypertension, Congestive Heart Failure, stroke TIA or Atrial Fibrillation
No record of antiplatelet or anticoagulent
No TIA in last month
Probability of thrombolysis in this hospital = 0.94

Create an embellished vignette for the patient, based on SSNAP data (this is performed manually).

Jan is a 75 year old retired accountant living in Bristol. Apart from being overweight, she considers herself fit and healthy and is enjoying being active in her retirement. On Wednesday, she visited her local shop at 11:30am to get a newspaper. While talking to Ben, the shop owner, she felt a sudden weakness in her right leg and fell to the floor. Ben found it difficult to get her up again as Jan had also lost some strength in her right arm. Noticing that Jan’s speech was slightly slurred, Ben remembered the ACT F.A.S.T. adverts on the TV and called the emergency services and an ambulance arrived 19 minutes later. The paramedics informed Jan it was likely she was experiencing a stroke and took her to hospital, where she arrived at 12:30pm.

Jan was immediately sent for a CT brain scan and by 1:00pm she was back in the emergency department being assessed by a doctor. The doctor found that the loss of movement on her right side meant that Jan could lift her arm and her leg above the horizontal but they did drift back onto the bed, and she had a very mild dysarthria. The stroke seemed relatively mild, scoring 4 on the NIH stroke scale. The CT scan showed no haemorrhage. The doctor thought that the prognosis for her mild lacunar stroke was favourable, and it did not justify the risks of thrombolysis. She admitted Jan to the stroke ward for observation and informed her that the consultant would come to see her the following morning.

Show tabular summary of SSNAP data, ignoring values which are zero.

patient_data = [d for d in confident.iloc[3].values if d!=0]
patient_index = [d for i,d in enumerate(confident.columns.values) 
                 if confident.iloc[3].values[i]!=0]

df = pd.DataFrame(data=patient_data, index = patient_index, columns = ['Patient'])

fig, ax = plt.subplots()
ax.axis('off')
ax.axis('tight')
t= ax.table(cellText=df.values, colWidths = [0.9]*len(df.columns),  rowLabels=df.index,  loc='center')
t.auto_set_font_size(False) 
t.set_fontsize(12)
#plt.tight_layout()
plt.show()

df.to_csv('./data/vignettes/confident_example.csv', index_label = 'Features')

Venn diagram for vignette hospital

plot_venn(data, 'MHMYL4920B', save=True)

Graphviz diagram for vignette hospital

hosp = 'MHMYL4920B'
plot_graph(data, hosp)
filename = f'./figures/vignettes/graph_{hosp}.gv.png'
Image(filename=filename)

Hospital with low treatment of patients with mild stroke#

Find a patient from a hospital that has a low thrombolysis rate in patients with minor strokes who had a minor stroke and wasn’t thrombolysed but benchmark hospitals would have.

Load table of previously calculated use and ratios by stroke severity and prior disability. This table provides use of thrombolysis in mild stroke (NIHSS 0-4) and moderate to sever stroke (NIHSS 16-20), and use in patients with no prior disability (modified Rankin Scale 0-1) and with prior disability (modified Rankin Scale 2+), and the rations between those groups.

ratios = pd.read_csv('./data/thrombolysis_ratios.csv')

ratios.head()

	StrokeTeam	minor	mod_severe	no_prior_disability	prior_disability	ratio_severity	ratio_disability
0	AGNOF1041H	0.171	0.463	0.398	0.235	0.370	0.589
1	AKCGO9726K	0.263	0.461	0.387	0.331	0.571	0.855
2	AOBTM3098N	0.029	0.337	0.246	0.157	0.087	0.640
3	APXEE8191H	0.070	0.341	0.263	0.140	0.205	0.534
4	ATDID5461S	0.156	0.290	0.296	0.131	0.539	0.442

minor_hospital = ratios.sort_values(by='minor')['StrokeTeam'].values[0]
ratios[ratios['StrokeTeam']==minor_hospital]

	StrokeTeam	minor	mod_severe	no_prior_disability	prior_disability	ratio_severity	ratio_disability
55	LFPMM4706C	0.006	0.152	0.069	0.047	0.041	0.675

The average use of thrombolysis in mild stroke (NIHSS 0-4) across hospitals is 12.9% of patients arriving within 4 hours of known stroke onset. This rate is, on average, 33% that of moderate to severe stroke (NIHSS 16-20).

LFPMM4706C is a stroke team that uses thrombolysis in 0.6% of mild stroke, 4% of the use rate in moderate to severe stroke in the same hospital.

minor = data.loc[(data['StrokeTeam'] == minor_hospital) & (data['S2NihssArrival']<=5) \
            & (data['Benchmark'] == 1) & (data['S2Thrombolysis']==0)]

minor

	StrokeTeam	S1AgeOnArrival	S1OnsetToArrival_min	S2RankinBeforeStroke	Visual	FacialPalsy	MotorArmLeft	MotorArmRight	MotorLegLeft	MotorLegRight	LimbAtaxia	Sensory	BestLanguage	Dysarthria	S2NihssArrival	S2BrainImagingTime_min	MoreEqual80y_No	MoreEqual80y_Yes	S1Gender_Female	S1Gender_Male	S1Ethnicity_White	S1OnsetInHospital_No	S1OnsetDateType_Precise	S1OnsetTimeType_Precise	S1ArriveByAmbulance_No	S1ArriveByAmbulance_Yes	S1AdmissionHour_06:00 to 08:59	S1AdmissionHour_09:00 to 11:59	S1AdmissionHour_12:00 to 14:59	S1AdmissionHour_15:00 to 17:59	S1AdmissionHour_18:00 to 20:59	S1AdmissionDay_Friday	S1AdmissionDay_Monday	S1AdmissionDay_Saturday	S1AdmissionDay_Sunday	S1AdmissionDay_Thursday	S1AdmissionDay_Tuesday	S1AdmissionQuarter_Q1	S1AdmissionQuarter_Q2	S1AdmissionQuarter_Q3	S1AdmissionYear_2016	S1AdmissionYear_2018	CongestiveHeartFailure_No	CongestiveHeartFailure_Yes	Hypertension_No	Hypertension_Yes	AtrialFibrillation_No	AtrialFibrillation_Yes	Diabetes_No	StrokeTIA_No	StrokeTIA_Yes	AFAntiplatelet_Yes	AFAntiplatelet_missing	AFAnticoagulent_No	AFAnticoagulent_missing	AFAnticoagulentVitK_missing	AFAnticoagulentDOAC_missing	AFAnticoagulentHeparin_missing	S2NewAFDiagnosis_missing	S2StrokeType_Infarction	S2TIAInLastMonth_No	S2TIAInLastMonth_missing	Benchmark	Predicted	Probability	AvgProbability
53644	LFPMM4706C	82.5	56.0	0	1.0	0.0	2.0	0.0	2.0	0.0	0.0	0.0	0.0	0.0	5.0	30.0	0	1	0	1	1	1	1	1	0	1	0	0	1	0	0	0	1	0	0	0	0	0	0	1	1	0	1	0	1	0	1	0	1	0	1	0	1	0	1	1	1	1	1	1	1	0	1	1	0.30	0.41
53656	LFPMM4706C	32.5	66.0	0	0.0	1.0	0.0	1.0	0.0	1.0	0.0	0.0	0.0	1.0	4.0	59.0	1	0	1	0	1	1	1	1	0	1	0	0	0	1	0	0	0	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.13	0.35
53674	LFPMM4706C	77.5	88.0	1	0.0	1.0	0.0	1.0	0.0	1.0	1.0	1.0	0.0	0.0	5.0	31.0	1	0	1	0	1	1	1	1	0	1	0	0	0	0	1	0	0	0	0	1	0	1	0	0	0	1	1	0	0	1	1	0	1	0	1	0	1	1	0	1	1	1	1	1	1	0	1	0	0.06	0.41
53769	LFPMM4706C	62.5	80.0	0	0.0	0.0	1.0	0.0	1.0	0.0	0.0	1.0	0.0	0.0	3.0	24.0	1	0	1	0	1	1	1	1	0	1	1	0	0	0	0	0	0	0	0	1	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.24	0.34
53823	LFPMM4706C	72.5	125.0	3	0.0	1.0	0.0	1.0	0.0	1.0	0.0	1.0	0.0	1.0	5.0	29.0	1	0	0	1	1	1	1	1	0	1	0	1	0	0	0	0	0	1	0	0	0	0	1	0	1	0	1	0	0	1	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	1	0.28	0.44
53828	LFPMM4706C	57.5	95.0	0	0.0	1.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	1.0	3.0	30.0	1	0	0	1	1	1	1	1	1	0	1	0	0	0	0	0	0	0	0	0	1	0	1	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.14	0.32
53838	LFPMM4706C	82.5	194.0	0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	0.0	1.0	1.0	3.0	16.0	0	1	0	1	1	1	1	1	0	1	0	0	1	0	0	0	0	0	1	0	0	1	0	0	0	1	1	0	1	0	1	0	1	1	0	0	1	1	0	1	1	1	1	1	0	1	1	0	0.02	0.28
53849	LFPMM4706C	82.5	53.0	0	1.0	0.0	0.0	2.0	0.0	2.0	0.0	0.0	0.0	0.0	5.0	29.0	0	1	1	0	1	1	1	1	0	1	0	0	0	1	0	1	0	0	0	0	0	0	1	0	1	0	0	1	0	1	0	1	1	1	0	1	0	1	0	1	1	1	1	1	0	1	1	1	0.31	0.36
53897	LFPMM4706C	72.5	68.0	0	0.0	0.0	1.0	0.0	2.0	2.0	0.0	0.0	0.0	0.0	5.0	24.0	1	0	0	1	1	1	1	1	0	1	1	0	0	0	0	0	0	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.21	0.52

Venn and GraphViz diagams for hospital with lowest thrombolysi use in mild strokes

plot_venn(data, minor_hospital, save=True)

plot_graph(data, minor_hospital)
filename = f'./figures/vignettes/graph_{minor_hospital}.gv.png'
Image(filename=filename)

Create vignette for a mild stroke patient at a hospital with low thrombolysis use in mild strokes

minor

	StrokeTeam	S1AgeOnArrival	S1OnsetToArrival_min	S2RankinBeforeStroke	Visual	FacialPalsy	MotorArmLeft	MotorArmRight	MotorLegLeft	MotorLegRight	LimbAtaxia	Sensory	BestLanguage	Dysarthria	S2NihssArrival	S2BrainImagingTime_min	MoreEqual80y_No	MoreEqual80y_Yes	S1Gender_Female	S1Gender_Male	S1Ethnicity_White	S1OnsetInHospital_No	S1OnsetDateType_Precise	S1OnsetTimeType_Precise	S1ArriveByAmbulance_No	S1ArriveByAmbulance_Yes	S1AdmissionHour_06:00 to 08:59	S1AdmissionHour_09:00 to 11:59	S1AdmissionHour_12:00 to 14:59	S1AdmissionHour_15:00 to 17:59	S1AdmissionHour_18:00 to 20:59	S1AdmissionDay_Friday	S1AdmissionDay_Monday	S1AdmissionDay_Saturday	S1AdmissionDay_Sunday	S1AdmissionDay_Thursday	S1AdmissionDay_Tuesday	S1AdmissionQuarter_Q1	S1AdmissionQuarter_Q2	S1AdmissionQuarter_Q3	S1AdmissionYear_2016	S1AdmissionYear_2018	CongestiveHeartFailure_No	CongestiveHeartFailure_Yes	Hypertension_No	Hypertension_Yes	AtrialFibrillation_No	AtrialFibrillation_Yes	Diabetes_No	StrokeTIA_No	StrokeTIA_Yes	AFAntiplatelet_Yes	AFAntiplatelet_missing	AFAnticoagulent_No	AFAnticoagulent_missing	AFAnticoagulentVitK_missing	AFAnticoagulentDOAC_missing	AFAnticoagulentHeparin_missing	S2NewAFDiagnosis_missing	S2StrokeType_Infarction	S2TIAInLastMonth_No	S2TIAInLastMonth_missing	Benchmark	Predicted	Probability	AvgProbability
53644	LFPMM4706C	82.5	56.0	0	1.0	0.0	2.0	0.0	2.0	0.0	0.0	0.0	0.0	0.0	5.0	30.0	0	1	0	1	1	1	1	1	0	1	0	0	1	0	0	0	1	0	0	0	0	0	0	1	1	0	1	0	1	0	1	0	1	0	1	0	1	0	1	1	1	1	1	1	1	0	1	1	0.30	0.41
53656	LFPMM4706C	32.5	66.0	0	0.0	1.0	0.0	1.0	0.0	1.0	0.0	0.0	0.0	1.0	4.0	59.0	1	0	1	0	1	1	1	1	0	1	0	0	0	1	0	0	0	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.13	0.35
53674	LFPMM4706C	77.5	88.0	1	0.0	1.0	0.0	1.0	0.0	1.0	1.0	1.0	0.0	0.0	5.0	31.0	1	0	1	0	1	1	1	1	0	1	0	0	0	0	1	0	0	0	0	1	0	1	0	0	0	1	1	0	0	1	1	0	1	0	1	0	1	1	0	1	1	1	1	1	1	0	1	0	0.06	0.41
53769	LFPMM4706C	62.5	80.0	0	0.0	0.0	1.0	0.0	1.0	0.0	0.0	1.0	0.0	0.0	3.0	24.0	1	0	1	0	1	1	1	1	0	1	1	0	0	0	0	0	0	0	0	1	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.24	0.34
53823	LFPMM4706C	72.5	125.0	3	0.0	1.0	0.0	1.0	0.0	1.0	0.0	1.0	0.0	1.0	5.0	29.0	1	0	0	1	1	1	1	1	0	1	0	1	0	0	0	0	0	1	0	0	0	0	1	0	1	0	1	0	0	1	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	1	0.28	0.44
53828	LFPMM4706C	57.5	95.0	0	0.0	1.0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	1.0	3.0	30.0	1	0	0	1	1	1	1	1	1	0	1	0	0	0	0	0	0	0	0	0	1	0	1	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.14	0.32
53838	LFPMM4706C	82.5	194.0	0	0.0	0.0	0.0	0.0	0.0	1.0	0.0	0.0	1.0	1.0	3.0	16.0	0	1	0	1	1	1	1	1	0	1	0	0	1	0	0	0	0	0	1	0	0	1	0	0	0	1	1	0	1	0	1	0	1	1	0	0	1	1	0	1	1	1	1	1	0	1	1	0	0.02	0.28
53849	LFPMM4706C	82.5	53.0	0	1.0	0.0	0.0	2.0	0.0	2.0	0.0	0.0	0.0	0.0	5.0	29.0	0	1	1	0	1	1	1	1	0	1	0	0	0	1	0	1	0	0	0	0	0	0	1	0	1	0	0	1	0	1	0	1	1	1	0	1	0	1	0	1	1	1	1	1	0	1	1	1	0.31	0.36
53897	LFPMM4706C	72.5	68.0	0	0.0	0.0	1.0	0.0	2.0	2.0	0.0	0.0	0.0	0.0	5.0	24.0	1	0	0	1	1	1	1	1	0	1	1	0	0	0	0	0	0	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	1	0	0.21	0.52

Patient at iloc 2 ( iloc 0 was predicted thrombolysed, iloc 1 is very young)

Age: 77.5
Onset to arrival: 88 mins
Rankin before: 1
Facial palsy: 1
Motor arm right: 1
Motor leg right: 1
Limb ataxia: 1
Sensory: 1
NIHSS: 5
Brain imaging time: 31 mins
White Female
Precise onset time
Admitted between 6 and 9pm on Thursday, quarter 1, 2018
No Diabetes, Congestive Heart Failure or Atrial Fibrillation
Yes to Stroke TIA and Hypertension
No record of antiplatelet or anticoagulent
No TIA in last month
Probability of thrombolysis in this hospital = 0!

Sally is a 76 year old woman with hypertension who had a TIA three years ago, who was taking clopidogrel, bendroflumethiazide and a statin. She found her knee arthritis a bit of a nuisance when shopping. She started to feel a slight weakness in her left arm and leg as she was making dinner for herself and her husband, Roy, one Thursday evening. She thought her knee was playing up and put the 6 O’Clock news on the radio to take her mind off it. However, when Sally and Roy sat down to eat at 7pm he noticed that she was clumsy with her fork and he thought her face looked twisted on the right. As she had previously had a TIA, Roy was alarmed and quickly dialled 999. An ambulance arrived within 10 minutes, and by 7:28 pm she was being assessed in the emergency department of her local hospital.

The first doctor to see Sally quickly sent her for a CT brain scan, which happened at 7:59pm. Once back in the emergency department, the doctor assessing Sally noted her history of hypertension and TIA but that she was otherwise generally well. The doctor noted a minor drooping of her face, and a drift of her left arm and leg which were also slightly numb, and it seemed to make her left arm rather clumsy. Her NIHSS was 5. The scan showed no signs of haemorrhage, but there was an old lacunar infarct in the left hemisphere. The doctor decided that the risks of thrombolysis outweighed her potential to benefit as she thought the natural prognosis from her mild stroke was good even without thrombolysis. She subsequently admitted her to the stroke unit for observation.

patient_data = [d for d in minor.iloc[2].values if d!=0]
patient_index = [d for i,d in enumerate(minor.columns.values) 
                 if minor.iloc[2].values[i]!=0]

df = pd.DataFrame(data=patient_data, index = patient_index, columns = ['Patient'])

fig, ax = plt.subplots()
ax.axis('off')
ax.axis('tight')
t= ax.table(cellText=df.values, colWidths = [0.9]*len(df.columns),  rowLabels=df.index,  loc='center')
t.auto_set_font_size(False) 
t.set_fontsize(12)
#plt.tight_layout()
plt.show()

df.to_csv('./data/vignettes/minor_example.csv', index_label = 'Features')

Hospital with low use of thrombolysis in patients with prior disability#

Example 1#

Find patients from a hospital that has a low rate of thrombolysis in patients with prior disability that:

had high disability
lost speech
wasn’t thrombolysed
benchmark would have thrombolysed

disability_hospital = ratios.sort_values(by='prior_disability')['StrokeTeam'].values[0]

disability = data.loc[(data['StrokeTeam'] == disability_hospital)\
                      & (data['S2RankinBeforeStroke']>=2) & (data['Dysarthria']>=2)\
            & (data['Benchmark'] == 1) & (data['S2Thrombolysis']==0)]

disability

	StrokeTeam	S1AgeOnArrival	S1OnsetToArrival_min	S2RankinBeforeStroke	Loc	LocQuestions	LocCommands	BestGaze	Visual	FacialPalsy	MotorArmLeft	MotorArmRight	MotorLegLeft	MotorLegRight	LimbAtaxia	Sensory	BestLanguage	Dysarthria	ExtinctionInattention	S2NihssArrival	S2BrainImagingTime_min	MoreEqual80y_Yes	S1Gender_Female	S1Ethnicity_White	S1OnsetInHospital_No	S1OnsetDateType_Precise	S1OnsetTimeType_Precise	S1ArriveByAmbulance_Yes	S1AdmissionHour_03:00 to 05:59	S1AdmissionHour_09:00 to 11:59	S1AdmissionHour_12:00 to 14:59	S1AdmissionHour_15:00 to 17:59	S1AdmissionDay_Friday	S1AdmissionDay_Saturday	S1AdmissionDay_Tuesday	S1AdmissionQuarter_Q1	S1AdmissionQuarter_Q2	S1AdmissionQuarter_Q4	S1AdmissionYear_2016	S1AdmissionYear_2017	CongestiveHeartFailure_No	Hypertension_No	Hypertension_Yes	AtrialFibrillation_No	AtrialFibrillation_Yes	Diabetes_No	Diabetes_Yes	StrokeTIA_No	StrokeTIA_Yes	AFAntiplatelet_No	AFAntiplatelet_missing	AFAnticoagulent_No	AFAnticoagulent_No but	AFAnticoagulent_missing	AFAnticoagulentVitK_missing	AFAnticoagulentDOAC_missing	AFAnticoagulentHeparin_missing	S2NewAFDiagnosis_missing	S2StrokeType_Infarction	S2TIAInLastMonth_No	S2TIAInLastMonth_missing	Benchmark	Majority_80	Probability	AvgProbability
75693	OUXUZ1084Q	87.5	110.0	3	0	2.0	2.0	0.0	1.0	3.0	0.0	1.0	0.0	1.0	0.0	0.0	3.0	2.0	2.0	17.0	37.0	1	1	1	1	1	1	1	0	0	1	0	0	0	1	1	0	0	0	1	1	1	0	1	0	0	1	1	0	0	1	0	0	1	1	1	1	1	1	0	1	1	0	0.25	0.46
75708	OUXUZ1084Q	82.5	151.0	3	0	1.0	1.0	0.0	2.0	2.0	0.0	1.0	0.0	1.0	0.0	1.0	2.0	2.0	0.0	13.0	9.0	1	1	1	1	1	1	1	0	0	0	1	0	0	1	1	0	0	0	1	1	0	1	1	0	1	0	1	0	0	1	0	0	1	1	1	1	1	1	0	1	1	1	0.22	0.56
75731	OUXUZ1084Q	97.5	108.0	2	1	2.0	0.0	2.0	0.0	1.0	4.0	0.0	4.0	0.0	2.0	0.0	2.0	2.0	0.0	20.0	29.0	1	1	1	1	1	1	1	0	0	1	0	0	1	0	0	1	0	0	1	1	0	1	0	1	1	0	1	0	1	0	0	1	0	1	1	1	1	1	0	1	1	0	0.15	0.36
75821	OUXUZ1084Q	82.5	80.0	3	1	2.0	2.0	1.0	1.0	1.0	0.0	2.0	0.0	4.0	0.0	1.0	3.0	2.0	2.0	22.0	30.0	1	1	1	1	1	1	1	1	0	0	0	0	1	0	1	0	0	0	1	1	0	1	0	1	1	0	1	0	1	0	1	0	0	1	1	1	1	1	0	1	1	0	0.23	0.45
75919	OUXUZ1084Q	92.5	122.0	3	0	2.0	2.0	1.0	3.0	1.0	0.0	4.0	0.0	3.0	0.0	2.0	3.0	2.0	2.0	25.0	28.0	1	1	1	1	1	1	1	0	1	0	0	0	1	0	0	1	0	0	1	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	0	0.15	0.35
75988	OUXUZ1084Q	92.5	94.0	3	3	2.0	2.0	2.0	3.0	3.0	3.0	3.0	3.0	3.0	0.0	2.0	3.0	2.0	2.0	36.0	13.0	1	1	1	1	1	1	1	0	0	1	0	1	0	0	0	0	1	1	0	1	0	1	1	0	0	1	1	0	0	1	0	0	1	1	1	1	1	1	0	1	1	0	0.23	0.41

The average use of thrombolysis in patients with prior disability (mRS 2+) is 18.9% of patients arriving within 4 hours of known stroke onset. This rate is, on average, 58% that of patients with no prior disability (mRS 0-1).

OUXUZ1084Q is a stroke team that uses thrombolysis in 2.3% of patients with prior disability, 10% of the use rate in patients with no prior disability.

plot_venn(data, disability_hospital, save=True)

plot_graph(data, disability_hospital)
filename = f'./figures/vignettes/graph_{disability_hospital}.gv.png'
Image(filename=filename)

disability

	StrokeTeam	S1AgeOnArrival	S1OnsetToArrival_min	S2RankinBeforeStroke	Loc	LocQuestions	LocCommands	BestGaze	Visual	FacialPalsy	MotorArmLeft	MotorArmRight	MotorLegLeft	MotorLegRight	LimbAtaxia	Sensory	BestLanguage	Dysarthria	ExtinctionInattention	S2NihssArrival	S2BrainImagingTime_min	MoreEqual80y_Yes	S1Gender_Female	S1Ethnicity_White	S1OnsetInHospital_No	S1OnsetDateType_Precise	S1OnsetTimeType_Precise	S1ArriveByAmbulance_Yes	S1AdmissionHour_03:00 to 05:59	S1AdmissionHour_09:00 to 11:59	S1AdmissionHour_12:00 to 14:59	S1AdmissionHour_15:00 to 17:59	S1AdmissionDay_Friday	S1AdmissionDay_Saturday	S1AdmissionDay_Tuesday	S1AdmissionQuarter_Q1	S1AdmissionQuarter_Q2	S1AdmissionQuarter_Q4	S1AdmissionYear_2016	S1AdmissionYear_2017	CongestiveHeartFailure_No	Hypertension_No	Hypertension_Yes	AtrialFibrillation_No	AtrialFibrillation_Yes	Diabetes_No	Diabetes_Yes	StrokeTIA_No	StrokeTIA_Yes	AFAntiplatelet_No	AFAntiplatelet_missing	AFAnticoagulent_No	AFAnticoagulent_No but	AFAnticoagulent_missing	AFAnticoagulentVitK_missing	AFAnticoagulentDOAC_missing	AFAnticoagulentHeparin_missing	S2NewAFDiagnosis_missing	S2StrokeType_Infarction	S2TIAInLastMonth_No	S2TIAInLastMonth_missing	Benchmark	Majority_80	Probability	AvgProbability
75693	OUXUZ1084Q	87.5	110.0	3	0	2.0	2.0	0.0	1.0	3.0	0.0	1.0	0.0	1.0	0.0	0.0	3.0	2.0	2.0	17.0	37.0	1	1	1	1	1	1	1	0	0	1	0	0	0	1	1	0	0	0	1	1	1	0	1	0	0	1	1	0	0	1	0	0	1	1	1	1	1	1	0	1	1	0	0.25	0.46
75708	OUXUZ1084Q	82.5	151.0	3	0	1.0	1.0	0.0	2.0	2.0	0.0	1.0	0.0	1.0	0.0	1.0	2.0	2.0	0.0	13.0	9.0	1	1	1	1	1	1	1	0	0	0	1	0	0	1	1	0	0	0	1	1	0	1	1	0	1	0	1	0	0	1	0	0	1	1	1	1	1	1	0	1	1	1	0.22	0.56
75731	OUXUZ1084Q	97.5	108.0	2	1	2.0	0.0	2.0	0.0	1.0	4.0	0.0	4.0	0.0	2.0	0.0	2.0	2.0	0.0	20.0	29.0	1	1	1	1	1	1	1	0	0	1	0	0	1	0	0	1	0	0	1	1	0	1	0	1	1	0	1	0	1	0	0	1	0	1	1	1	1	1	0	1	1	0	0.15	0.36
75821	OUXUZ1084Q	82.5	80.0	3	1	2.0	2.0	1.0	1.0	1.0	0.0	2.0	0.0	4.0	0.0	1.0	3.0	2.0	2.0	22.0	30.0	1	1	1	1	1	1	1	1	0	0	0	0	1	0	1	0	0	0	1	1	0	1	0	1	1	0	1	0	1	0	1	0	0	1	1	1	1	1	0	1	1	0	0.23	0.45
75919	OUXUZ1084Q	92.5	122.0	3	0	2.0	2.0	1.0	3.0	1.0	0.0	4.0	0.0	3.0	0.0	2.0	3.0	2.0	2.0	25.0	28.0	1	1	1	1	1	1	1	0	1	0	0	0	1	0	0	1	0	0	1	1	0	1	0	1	0	1	0	1	1	0	0	1	0	1	1	1	1	1	1	0	1	0	0.15	0.35
75988	OUXUZ1084Q	92.5	94.0	3	3	2.0	2.0	2.0	3.0	3.0	3.0	3.0	3.0	3.0	0.0	2.0	3.0	2.0	2.0	36.0	13.0	1	1	1	1	1	1	1	0	0	1	0	1	0	0	0	0	1	1	0	1	0	1	1	0	0	1	1	0	0	1	0	0	1	1	1	1	1	1	0	1	1	0	0.23	0.41

Patient at iloc 1 (benchmark and 80% majority would thrombolyse)

Age: 82.5
Onset to arrival: 151 mins
Rankin before: 3
LocQuestions: 1
LocCommands: 1
Visual: 2
Facial palsy: 2
Motor arm right: 1
Motor leg right: 1
Sensory: 1
BestLanguage: 2
Dysarthia: 2
NIHSS: 13
Brain imaging time: 9 mins
White Female
Precise onset time
Admitted between 3 and 6pm on Tuesday, quarter 1, 2017
No Diabetes, Congestive Heart Failure, StrokeTIA or Atrial Fibrillation
Yes to Hypertension
No record of antiplatelet, anticoagulent or newAFdiagnosis
No recorded TIA in last month
Probability of thrombolysis in this hospital = 0.22

Mabel is 84 years old and, since her husband passed away, has lived in a residential home: although she is still mobile, Mabel requires some help with her day to day living due to memory decline. After finishing her lunch at 1pm on Tuesday, Katy, one of the staff, came over to take her plate to the kitchen. Katy always enjoyed a chat with Mabel however, as they were talking, Mabel started to slur her words and very quickly it became hard for Katy to understand what she was saying. Knowing the signs of stroke, Katy quickly called for an ambulance.

Unfortunately, due to a serious incident nearby, the ambulance almost 2 hours to arrive. Once it did arrive, Mabel was quickly taken to the nearest stroke unit. She arrived at 3:30pm and had a brain scan within 10 minutes. The clinician assessing her found her stroke was moderate, scoring 13 on the NIH stroke scale. Mabel had previously been diagnosed with hypertension but was otherwise healthy for her age. Mabel was admitted to the stroke unit but was not offered thrombolysis.

patient_data = [d for d in disability.iloc[1].values if d!=0]
patient_index = [d for i,d in enumerate(disability.columns.values) 
                 if disability.iloc[1].values[i]!=0]

df = pd.DataFrame(data=patient_data, index = patient_index, columns = ['Patient'])

fig, ax = plt.subplots()
ax.axis('off')
ax.axis('tight')
t= ax.table(cellText=df.values, colWidths = [0.9]*len(df.columns),  rowLabels=df.index,  loc='center')
t.auto_set_font_size(False) 
t.set_fontsize(12)
#plt.tight_layout()
plt.show()

df.to_csv('./data/vignettes/disability_example.csv', index_label = 'Features')

Example 2#

Rankin = 4 and NIHSS < 10

disability = data.loc[(data['S2RankinBeforeStroke']>=4) & (data['Dysarthria']>=2)\
            & (data['Benchmark'] == 1) & (data['S2Thrombolysis']==0)\
                     & (data['S2NihssArrival']<10) & (data['Predicted']==0)]#['Dysarthria']

disability

	StrokeTeam	S1AgeOnArrival	S1OnsetToArrival_min	S2RankinBeforeStroke	Loc	LocQuestions	LocCommands	BestGaze	Visual	FacialPalsy	MotorArmRight	MotorLegLeft	MotorLegRight	LimbAtaxia	Sensory	BestLanguage	Dysarthria	ExtinctionInattention	S2NihssArrival	S2BrainImagingTime_min	MoreEqual80y_No	MoreEqual80y_Yes	S1Gender_Female	S1Gender_Male	S1Ethnicity_Other	S1Ethnicity_White	S1OnsetInHospital_No	S1OnsetDateType_Precise	S1OnsetTimeType_Best estimate	S1OnsetTimeType_Precise	S1ArriveByAmbulance_Yes	S1AdmissionHour_09:00 to 11:59	S1AdmissionHour_12:00 to 14:59	S1AdmissionHour_15:00 to 17:59	S1AdmissionHour_18:00 to 20:59	S1AdmissionHour_21:00 to 23:59	S1AdmissionDay_Monday	S1AdmissionDay_Saturday	S1AdmissionDay_Sunday	S1AdmissionDay_Thursday	S1AdmissionDay_Tuesday	S1AdmissionDay_Wednesday	S1AdmissionQuarter_Q1	S1AdmissionQuarter_Q2	S1AdmissionQuarter_Q3	S1AdmissionQuarter_Q4	S1AdmissionYear_2016	S1AdmissionYear_2017	S1AdmissionYear_2018	CongestiveHeartFailure_No	Hypertension_No	Hypertension_Yes	AtrialFibrillation_No	AtrialFibrillation_Yes	Diabetes_No	Diabetes_Yes	StrokeTIA_No	StrokeTIA_Yes	AFAntiplatelet_No	AFAntiplatelet_Yes	AFAntiplatelet_missing	AFAnticoagulent_No	AFAnticoagulent_No but	AFAnticoagulent_missing	AFAnticoagulentVitK_No	AFAnticoagulentVitK_missing	AFAnticoagulentDOAC_No	AFAnticoagulentDOAC_missing	AFAnticoagulentHeparin_No	AFAnticoagulentHeparin_missing	S2NewAFDiagnosis_No	S2NewAFDiagnosis_missing	S2StrokeType_Infarction	S2TIAInLastMonth_No	S2TIAInLastMonth_missing	Benchmark	Probability	AvgProbability
6244	JINXD0311F	97.5	71.0	4	0	2.0	1.0	0.0	1.0	1.0	0.0	0.0	0.0	0.0	0.0	2.0	2.0	0.0	9.0	19.0	0	1	1	0	1	0	1	1	0	1	1	0	1	0	0	0	0	1	0	0	0	0	0	0	0	1	0	0	1	1	0	1	1	0	0	1	1	0	0	0	1	0	1	0	1	0	1	0	1	0	1	0	1	0	1	1	0.39	0.43
9961	AGNOF1041H	87.5	79.0	4	0	0.0	1.0	0.0	0.0	1.0	0.0	0.0	0.0	0.0	1.0	0.0	2.0	0.0	5.0	4.0	0	1	1	0	0	1	1	1	0	1	1	1	0	0	0	0	1	0	0	0	0	0	0	1	0	0	0	1	0	1	1	0	1	0	0	1	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	1	1	0.30	0.40
19408	QNARI2373R	97.5	188.0	4	0	0.0	0.0	0.0	0.0	1.0	3.0	0.0	3.0	0.0	0.0	0.0	2.0	0.0	9.0	30.0	0	1	1	0	0	1	1	1	0	1	1	0	1	0	0	0	1	0	0	0	0	0	0	1	0	0	0	0	1	1	0	1	1	0	1	0	1	0	0	0	1	1	0	0	1	0	1	0	1	0	1	0	1	0	1	1	0.29	0.35
22320	ROOIZ9592Y	82.5	179.0	4	0	0.0	1.0	0.0	0.0	1.0	0.0	0.0	0.0	0.0	0.0	3.0	2.0	0.0	7.0	12.0	0	1	0	1	0	1	1	1	0	1	1	0	1	0	0	0	0	0	1	0	0	0	0	0	1	0	0	1	0	1	0	1	1	0	1	0	0	1	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	1	1	0.40	0.38
23496	ISIZF6614O	57.5	109.0	4	1	2.0	2.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	2.0	0.0	7.0	13.0	1	0	0	1	0	1	1	1	0	1	1	0	0	0	0	1	0	0	0	0	0	1	0	0	0	1	0	1	0	1	1	0	1	0	0	1	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	1	1	0.39	0.41
33217	UIWEN7236N	82.5	79.0	4	0	0.0	0.0	0.0	2.0	1.0	0.0	0.0	0.0	1.0	1.0	1.0	2.0	1.0	9.0	27.0	0	1	1	0	0	1	1	1	1	0	1	1	0	0	0	0	0	0	0	1	0	0	1	0	0	0	0	0	1	1	1	0	1	0	1	0	1	0	0	0	1	1	0	0	1	0	1	0	1	0	1	0	1	0	1	1	0.33	0.31
34529	JADBS8258F	72.5	73.0	4	0	2.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	2.0	2.0	0.0	6.0	22.0	1	0	1	0	0	1	1	1	0	1	1	1	0	0	0	0	0	0	0	0	0	1	0	0	1	0	0	0	1	1	1	0	0	1	1	0	0	1	1	0	0	0	1	0	1	0	1	0	1	0	0	1	1	0	1	1	0.31	0.33
40023	FAJKD7118X	72.5	59.0	4	0	0.0	0.0	0.0	0.0	1.0	0.0	2.0	0.0	0.0	0.0	0.0	2.0	0.0	5.0	20.0	1	0	0	1	0	1	1	1	0	1	1	0	1	0	0	0	0	0	0	1	0	0	0	0	1	0	1	0	0	1	1	0	1	0	0	1	0	1	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	1	0	1	0.24	0.39
59776	LZGVG8257A	87.5	164.0	4	0	1.0	0.0	2.0	0.0	2.0	0.0	0.0	0.0	0.0	0.0	0.0	2.0	2.0	9.0	29.0	0	1	0	1	0	1	1	1	0	1	1	0	0	0	1	0	0	0	1	0	0	0	0	0	1	0	0	0	1	1	0	1	1	0	0	1	1	0	0	0	1	1	0	0	1	0	1	0	1	0	1	0	1	0	1	1	0.33	0.41
80168	JHDQL1362V	97.5	53.0	4	2	0.0	0.0	2.0	0.0	1.0	0.0	0.0	0.0	0.0	0.0	2.0	2.0	0.0	9.0	14.0	0	1	1	0	0	1	1	1	0	1	1	1	0	0	0	0	0	0	0	0	1	0	0	0	1	0	0	0	1	1	1	0	1	0	1	0	1	0	0	0	1	1	0	0	1	0	1	0	1	0	1	0	1	0	1	1	0.41	0.43
82615	VZZSN1825O	87.5	48.0	5	0	2.0	0.0	0.0	0.0	0.0	1.0	0.0	1.0	0.0	0.0	2.0	2.0	0.0	8.0	18.0	0	1	0	1	0	1	1	1	0	1	1	0	0	1	0	0	0	0	0	1	0	0	0	0	0	1	1	0	0	1	0	1	1	0	1	0	1	0	0	0	1	0	0	1	0	1	0	1	0	1	0	1	1	0	1	1	0.36	0.48
86187	BICAW1125K	82.5	117.0	4	1	1.0	0.0	2.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	1.0	2.0	1.0	8.0	21.0	0	1	1	0	0	1	1	1	0	1	1	0	0	1	0	0	0	0	1	0	0	0	0	0	0	1	0	1	0	1	1	0	0	1	1	0	1	0	0	1	0	1	0	0	0	1	0	1	0	1	0	1	1	0	1	1	0.33	0.37

Patient at iloc -1#

Age: 82.5
Onset to arrival: 117 mins
Rankin before: 4
Loc: 1
LocQuestions: 1
Best Gaze: 2
BestLanguage: 1
Dysarthia: 2
Extinction Inattention: 1
NIHSS: 8
Brain imaging time: 21 mins
White Female
Precise onset time
Admitted between 3 and 6pm on Sunday, quarter 4, 2017
No Diabetes, Congestive Heart Failure, StrokeTIA or Hypertension
Yes to Atrial Fibrillation and AF Anit Platelet
No anticoagulent or newAFdiagnosis
No recorded TIA in last month
Probability of thrombolysis in this hospital = 0.33

Mabel is 84 years old and, since her husband passed away, has lived in a residential home: although she is still mobile with assistance, Mabel requires help with her day to day activities due to her memory problems, and couldn’t be left on her own for any length of time. After finishing her lunch at 1:30pm one Sunday, Katy, one of the staff, helped her to walk back to her room. When Katy went back to check on Mabel an hour later, it was very hard for Katy to understand what she was saying. Knowing the signs of stroke, Katy quickly called for an ambulance, which arrived 20 minutes later.

Mabel arrived at her local hospital at 3:27pm and had a CT brain scan within 25 minutes. The clinician assessing her found her a bit drowsy and disorientated, struggling to speak clearly, with her attention less readily drawn to the left side but no limb weakness, scoring 8 on the NIH stroke scale although this seemed to be overstated by her dementia. Mabel’s CT scan showed significant cerebral atrophy with confluent symmetrical hypodensities suggesting an advanced vascular dementia. Given her severe pre-existing disability and the advanced vascular changes on her CT scan, the admitting doctor considered the risks of haemorrhage to be high, and her chances of responding to thrombolysis with a meaningful reduction in her disability to be low, and so she admitted her to the stroke unit directly without thrombolysis.

patient_data = [d for d in disability.iloc[-1].values if d!=0]
patient_index = [d for i,d in enumerate(disability.columns.values) 
                 if disability.iloc[-1].values[i]!=0]

df = pd.DataFrame(data=patient_data, index = patient_index, columns = ['Patient'])

df.to_csv('./data/vignettes/disability_example_2.csv', index_label = 'Features')

plot_venn(data, 'BICAW1125K', save=True)

hosp = 'BICAW1125K'
plot_graph(data, hosp)
filename = f'./figures/vignettes/graph_{hosp}.gv.png'
Image(filename=filename)

Observations#

We illustrate methods of visualisation of use of thrombolysis compared with benchmark hospitals.
We can identify hospitals with a pre-determined difference in use of thrombolysis, and compare to other hospitals though hospital-level diagrams or patient vignettes.
Feedback during workshops with physicians was that the Venn diagrams were difficult to understand; we then added the Graph diagrams as a result
The synthetic patient vignettes were popular as a discussion tool at the workshop with physicians.

SAMueL Stroke Audit Machine Learning 1

Thrombolysis hospital vs. benchmark diagrams and patient vignettes

Contents