def format_nct (nct):
snct = []
url = 'http://clinicaltrials.gov/ct2/show/%s?displayxml=true'
for ct in nct:
if ct[0] in app.jinja_env.globals['ctgov_idx']:
title = app.jinja_env.globals['ctgov_idx'][ct[0]][0]
cond = format_condition(app.jinja_env.globals['ctgov_idx'][ct[0]][1])
else:
xmltree = ElementTree.fromstring (download_web_data(url % ct[0]))
title = xmltree.find ('brief_title')
# title
if title is None:
title = ct[0]
else:
title = title.text
# condition
lcond = xmltree.findall ('condition')
cond = ''
if lcond is not None:
for c in lcond:
if c is not None:
cond = '%s; %s' % (cond, c.text)
cond = cond[2:]
app.jinja_env.globals['ctgov_idx'][ct[0]] = (title, cond)
cond = format_condition (cond)
log.info ('added %s to the CTGOV idx (updated to %d trials)' % (ct[0], len(app.jinja_env.globals['ctgov_idx'])))
snct.append ((ct[0], ct[1], title, cond))
return snct
def get_nct_number(url):
xml = download_web_data(url)
if xml is None:
return 0
xmltree = ElementTree.fromstring(xml)
nnct = xmltree.get('count')
return int(nnct)
def format_nct(nct):
snct = []
url = 'http://clinicaltrials.gov/ct2/show/%s?displayxml=true'
for ct in nct:
if ct[0] in app.jinja_env.globals['ctgov_idx']:
title = app.jinja_env.globals['ctgov_idx'][ct[0]][0]
cond = format_condition(
app.jinja_env.globals['ctgov_idx'][ct[0]][1])
else:
xmltree = ElementTree.fromstring(download_web_data(url % ct[0]))
title = xmltree.find('brief_title')
# title
if title is None:
title = ct[0]
else:
title = title.text
# condition
lcond = xmltree.findall('condition')
cond = ''
if lcond is not None:
for c in lcond:
if c is not None:
cond = '%s; %s' % (cond, c.text)
cond = cond[2:]
app.jinja_env.globals['ctgov_idx'][ct[0]] = (title, cond)
cond = format_condition(cond)
log.info('added %s to the CTGOV idx (updated to %d trials)' %
(ct[0], len(app.jinja_env.globals['ctgov_idx'])))
snct.append((ct[0], ct[1], title, cond))
return snct
def get_initial_nct_from_url_ad(url):
# num. of studies available
'''
n = get_nct_number('%s&count=0' % url)
if n == 0:
return []
# get the list of clinical studies
xmltree = ElementTree.fromstring (download_web_data('%s&count=%d' % (url, n)))
lnct = xmltree.findall ('clinical_study')
rnct = []
i = 1
for ct in lnct:
ids = ct.find ('nct_id')
if ids is None:
continue
rnct.append ('%s;%d' % (ids.text,i))
i += 1
'''
n = 2000
xmltree = ElementTree.fromstring(
download_web_data('%s&count=%d' % (url, n)))
rnct = []
i = 1
for ct in xmltree.iter('nct_id'):
ids = ct
if ids is None:
continue
rnct.append('%s;%d' % (ids.text, i))
i += 1
return rnct
def get_nct_number (url):
xml = download_web_data(url)
if xml is None:
return 0
xmltree = ElementTree.fromstring (xml)
nnct = xmltree.get ('count')
return int(nnct)
def get_disease_clinical_trials (ldisease): #retrieve Trial IDs from clincialtrials.gov
disease_to_nct = {}
stat = []
#ldisease = sorted(map(lambda x:' '.join(x.lower().split()), ldisease))
trial_ids = []
i=1
d = ldisease.replace (',', '')
fd = d.replace(' ', '+')
url = 'https://clinicaltrials.gov/ct2/results?cond=%s&displayxml=true'
# num. of studies available
if ldisease=='all':
url='https://clinicaltrials.gov/ct2/results?%sdisplayxml=true'
fd=''
print url
xml = download_web_data(url % (fd))
xmltree = xml_parser.fromstring (xml)
n = int(xmltree.get ('count'))
print n
i=i+1
nct = set()
url_final = url + '&start=%d&count=%s';
for j in range(1, n, 1):
if j % 5000 ==0:
print "id: 0-",j," ..."
xmltree = xml_parser.fromstring (download_web_data(url_final % (fd, j, 1)))
lnct = xmltree.findall ('clinical_study')
for ct in lnct:
cod = ct.find ('nct_id')
if cod is None:
continue
#print "trial %s"%cod.text
trial_ids.append((cod.text))
return trial_ids
def retrieve_trials(url, npag):
# num. of studies available
n = of.format_nct_number(get_nct_number('%s&count=0' % url))
# get the list of clinical studies
xml = download_web_data('%s&pg=%s' % (url, npag))
if xml is None:
return (0, [])
xmltree = ElementTree.fromstring(xml)
lnct = xmltree.findall('clinical_study')
nct = []
for ct in lnct:
pct = parse_xml_nct(ct)
if pct[0] is not None:
cond = of.format_condition(pct[3])
nct.append((pct[0], pct[1], pct[2], cond))
return (n, nct)
def get_initial_nct_from_url (url):
# num. of studies available
n = get_nct_number('%s&count=0' % url)
if n == 0:
return []
# get the list of clinical studies
xmltree = ElementTree.fromstring (download_web_data('%s&count=%d' % (url, n)))
lnct = xmltree.findall ('clinical_study')
rnct = []
i = 1
for ct in lnct:
ids = ct.find ('nct_id')
if ids is None:
continue
rnct.append ('%s;%d' % (ids.text,i))
i += 1
return rnct
def retrieve_trials (url, npag):
# num. of studies available
n = of.format_nct_number(get_nct_number('%s&count=0' % url))
# get the list of clinical studies
xml = download_web_data('%s&pg=%s' % (url, npag))
if xml is None:
return (0, [])
xmltree = ElementTree.fromstring (xml)
lnct = xmltree.findall ('clinical_study')
nct = []
for ct in lnct:
pct = parse_xml_nct (ct)
if pct[0] is not None:
if pct[0] not in app.jinja_env.globals['ctgov_idx']:
app.jinja_env.globals['ctgov_idx'][pct[0]] = (pct[2], pct[3])
cond = of.format_condition (pct[3])
nct.append ((pct[0], pct[1], pct[2], cond))
return (n, nct)
def extract_criteria(cid): #Using IDs to retrieve eligibility criteria
output = ""
if cid is not None:
url_trial = 'http://clinicaltrials.gov/show/%s?displayxml=true'
#url_trial ='http://clinicaltrials.gov/search?term=%s&displayxml=true'
page = download_web_data(url_trial % cid)
#with codecs.open('temp.txt', 'w','utf8') as writer:
# writer.write(page)
#with codec.open('temp.txt', 'r', 'utf8') as reader:
if page is not None:
ct_xml = xml_parser.fromstring (page)
ec = ct_xml.find ('eligibility')
if ec is not None:
# parse to get criteria text
d = ec.find ('criteria')
if d is not None:
txt = d.find ('textblock')
if txt is not None:
output = txt.text
return output
def extract_description(cid): #Using IDs to retrieve eligibility criteria
output = ""
if cid is not None:
url_trial = 'http://clinicaltrials.gov/show/%s?displayxml=true'
#url_trial ='http://clinicaltrials.gov/search?term=%s&displayxml=true'
page = download_web_data(url_trial % cid)
#with codecs.open('temp.txt', 'w','utf8') as writer:
# writer.write(page)
#with codec.open('temp.txt', 'r', 'utf8') as reader:
if page is not None:
ct_xml = xml_parser.fromstring (page)
summary = ct_xml.find ('brief_summary')
if summary is not None:
txt = summary.find ('textblock')
if txt is not None:
output = txt.text
description = ct_xml.find('detailed_description')
if description is not None:
txt2 = summary.find('textblock')
if txt2 is not None:
output = output+txt2.text
return output
