Purge 4 misattributed publication entries from YAMLs

Three PMIDs were returned by PubMed's "Westover MB[Author]" query but
do not actually belong to Brandon Westover (NCBI author disambiguation
catches a different M. Westover or unrelated co-authors):

  37245479 — Einizade et al, ProductGraphSleepNet (no Westover author)
  36131149 — Huang et al, AI foundation for therapeutic science
  34889311 — Kanth et al, Cancer Risk in Serrated Polyposis (different M Westover)

Adds an EXCLUDE_PMIDS set to scrape_pubmed_auto.py so the nightly auto-
scraper drops these on parse and they don't reappear.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: mb-westover

_data/yamlML_AI.yml

@@ -1015,16 +1015,6 @@
     display: 'JMIR AI 2023 Jan-Dec; 2 (1) doi: 10.2196/46769. Epub 2023 Jul 26. PMID:38090533
       ; PMCID:PMC10715804 '
   highlight: 0
-- title: '"ProductGraphSleepNet: Sleep staging using product spatio-temporal graph
-    learning with attentive temporal aggregation."'
-  image: ''
-  description: ''
-  authors: Einizade A, Nasiri S, Sardouie SH, Clifford GD.
-  link:
-    url: https://pubmed.ncbi.nlm.nih.gov/37245479
-    display: 'Neural Netw 2023 Jul; 164 :667-680 doi: 10.1016/j.neunet.2023.05.016.
-      Epub 2023 May 13. PMID:37245479 '
-  highlight: 0
 - title: '"Identification of patients with epilepsy using automated electronic health
     records phenotyping."'
   image: ''
@@ -1097,16 +1087,6 @@
     display: 'Neuroinformatics 2022 Oct; 20 (4) :943-964 doi: 10.1007/s12021-022-09572-9.
       Epub 2022 Mar 28. PMID:35347570 ; PMCID:PMC9515245 '
   highlight: 0
-- title: '"Artificial intelligence foundation for therapeutic science."'
-  image: ''
-  description: ''
-  authors: Huang K, Fu T, Gao W, Zhao Y, Roohani Y, Leskovec J, Coley CW, Xiao C,
-    Sun J, Zitnik M.
-  link:
-    url: https://pubmed.ncbi.nlm.nih.gov/36131149
-    display: 'Nat Chem Biol 2022 Oct; 18 (10) :1033-1036 doi: 10.1038/s41589-022-01131-2.
-      PMID:36131149 ; PMCID:PMC9529840 '
-  highlight: 0
 - title: '"Automated Scoring of Respiratory Events in Sleep With a Single Effort Belt
     and Deep Neural Networks."'
   image: ''

_data/yamlriskPrediction.yml

@@ -196,17 +196,6 @@
     display: 'Urol Oncol 2022 Apr; 40 (4) :161.e1-161.e7 doi: 10.1016/j.urolonc.2021.08.007.
       Epub 2021 Aug 29. PMID:34465541 ; PMCID:PMC8882704 '
   highlight: 0
-- title: '"Cancer Risk in Patients With and Relatives of Serrated Polyposis Syndrome
-    and Sporadic Sessile Serrated Lesions."'
-  image: ''
-  description: ''
-  authors: Kanth P, Yu Z, Keener MB, Koptiuch C, Kohlmann WK, Neklason DW, Westover
-    M, Curtin K.
-  link:
-    url: https://pubmed.ncbi.nlm.nih.gov/34889311
-    display: 'Am J Gastroenterol 2022 Feb 1; 117 (2) :336-342 doi: 10.14309/ajg.0000000000001572.
-      PMID:34889311 '
-  highlight: 0
 - title: '"Automated Annotation of Epileptiform Burden and Its Association with Outcomes."'
   image: ''
   description: ''

_data/yamlsleepStaging.yml

@@ -345,16 +345,6 @@
     display: 'JMIR AI 2023 Jan-Dec; 2 (1) doi: 10.2196/46769. Epub 2023 Jul 26. PMID:38090533
       ; PMCID:PMC10715804 '
   highlight: 0
-- title: '"ProductGraphSleepNet: Sleep staging using product spatio-temporal graph
-    learning with attentive temporal aggregation."'
-  image: ''
-  description: ''
-  authors: Einizade A, Nasiri S, Sardouie SH, Clifford GD.
-  link:
-    url: https://pubmed.ncbi.nlm.nih.gov/37245479
-    display: 'Neural Netw 2023 Jul; 164 :667-680 doi: 10.1016/j.neunet.2023.05.016.
-      Epub 2023 May 13. PMID:37245479 '
-  highlight: 0
 - title: '"Exploiting labels from multiple experts in automated sleep scoring."'
   image: ''
   description: ''

scrape_pubmed_auto.py

@@ -24,6 +24,15 @@
 CDAC_DOWNLOADS_BASE = "https://bdsp-core.github.io/cdac-downloads/"
 PDF_PMID_RE = re.compile(r"_(\d{7,9})\.pdf$", re.IGNORECASE)
 
+# PMIDs that PubMed's "Westover MB[Author]" search returns but do not actually
+# belong to Brandon Westover (different M. Westover, ambiguous author records,
+# etc.). Confirmed with the PI; never add these to the publication YAMLs.
+EXCLUDE_PMIDS = {
+    "37245479",  # Einizade et al, ProductGraphSleepNet (no Westover author)
+    "36131149",  # Huang et al, AI foundation for therapeutic science (no Westover)
+    "34889311",  # Kanth et al, Cancer Risk in Serrated Polyposis (different M Westover)
+}
+
 
 def fetch_pdf_pmid_map():
     """Return {PMID: download_url} for PDFs in the cdac-downloads repo whose
@@ -159,7 +168,7 @@ def fetch_publication_details(self, pmids):
                 # Parse publications
                 for article in root.findall('.//PubmedArticle'):
                     pub = self.parse_article(article)
-                    if pub:
+                    if pub and pub.get('pmid') not in EXCLUDE_PMIDS:
                         all_publications.append(pub)
 
                 # Rate limiting