ASD Traits Analysis and Early Indicators
Autistic Spectrum Disorder Screening Data for Toddlers
Addressing the urgent global increase in Autism Spectrum Disorder (ASD), this initiative proposes a new, behavior-focused dataset aimed at enhancing early ASD screening and diagnosis through efficient and accessible methods.
In this dataset, we record ten behavioural features (Q-Chat-10) plus other individuals characteristics that have proved to be effective in detecting the ASD cases from controls in behaviour science.
import pandas as pd
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
data = pd.read_csv("Toddler Autism dataset July 2018.csv") data
labels = ["Yes", "No"] myexplode = [0.1, 0] plt.pie(data.groupby('Class/ASD Traits ').size().reindex(labels),explode = myexplode, shadow = True,autopct='%1.1f%%') plt.legend(title = "ASD:",labels = labels) plt.show()
sns.set_style("whitegrid")
as_yes_responses = data[data['Class/ASD Traits '] == 'Yes'][[f'A{i}' for i in range(1, 11)]].sum()
as_no_responses = data[data['Class/ASD Traits '] == 'No'][[f'A{i}' for i in range(1, 11)]].sum()
responses_df = pd.DataFrame({
'Question': [f'A{i}' for i in range(1, 11)],
'Yes - ASD Traits': as_yes_responses.values,
'No - ASD Traits': as_no_responses.values
}).melt(id_vars="Question", var_name="ASD Traits Presence", value_name="Frequency")
plt.figure(figsize=(14, 8))
sns.barplot(x='Question', y='Frequency', hue='ASD Traits Presence', data=responses_df)
plt.title('Frequency of "Yes" Responses for Each Question by ASD Traits Presence')
plt.xlabel('Question')
plt.ylabel('Frequency of "Yes" Responses')
plt.legend(title='ASD Traits Presence', loc='upper right')
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()
Dr. Fayez Thabtah collected asked parents of 1054 toddlers of age less than three years old 10 Yes/No behavioural questions. The number of "Yes" were added up and have been recorded in the column called "Q chat 10 question". If this column has more than 3 "Yes", toddlers will be classified as positive ASD
age_specific_responses_corrected = data[data['Class/ASD Traits '] == 'Yes'].groupby('Age_Mons')[['A1', 'A4', 'A7']].mean()
plt.figure(figsize=(14, 8))
for question in ['A1', 'A4', 'A7']:
plt.plot(age_specific_responses_corrected.index, age_specific_responses_corrected[question], marker='o', label=f'Question {question}')
plt.title('Age-specific Trends in "Yes" Responses to Selected Questions for Toddlers with ASD Traits')
plt.xlabel('Age in Months')
plt.ylabel('Proportion of "Yes" Responses')
plt.legend(title='Question')
plt.grid(True)
plt.savefig('trends.png', dpi=300)
plt.tight_layout()
plt.show()
labels = ['middle eastern', 'White European', 'Hispanic', 'black', 'asian',
'south asian', 'Native Indian', 'Others', 'Latino', 'mixed',
'Pacifica']
plt.figure(figsize=(10, 6))
plt.pie(data.groupby('Ethnicity').size().reindex(labels), shadow = True,autopct='%1.1f%%')
plt.legend(title = "Ethnicity:",labels = labels,loc='center left', bbox_to_anchor=(1, 0.5))
plt.savefig('race.png', dpi=300)
plt.show()
labels = ['middle eastern', 'White European', 'Hispanic', 'black', 'asian',
'south asian', 'Native Indian', 'Others', 'Latino', 'mixed',
'Pacifica']
plt.figure(figsize=(10, 6))
plt.pie(data.groupby('Ethnicity').size().reindex(labels), shadow = True,autopct='%1.1f%%')
plt.legend(title = "Ethnicity:",labels = labels,loc='center left', bbox_to_anchor=(1, 0.5))
plt.savefig('race.png', dpi=300)
plt.show()
1) Tabtah, F. (2017). Autism Spectrum Disorder Screening: Machine Learning Adaptation and DSM-5 Fulfillment. Proceedings of the 1st International Conference on Medical and Health Informatics 2017, pp.1-6. Taichung City, Taiwan, ACM.
2) Thabtah, F. (2017). ASDTests. A mobile app for ASD screening. www.asdtests.com [accessed December 20th, 2017].
3) Thabtah, F. (2017). Machine Learning in Autistic Spectrum Disorder Behavioural Research: A Review. Informatics for Health and Social Care Journal.
4) Thabtah F, Kamalov F., Rajab K (2018) A new computational intelligence approach to detect autistic features for autism screening. International Journal of Medical Infromatics, Volume 117, pp. 112-124.