Bokai-Zhao

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Cross-modal co-attention Transformer fusion for histology-genomics survival prediction. Predecessor of PORPOISE in Mahmood Lab's multimodal pathology line.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

The canonical reference for slide-level zero-shot pathology VL via top-k aggregation. Cite when slide-level zero-shot WSI classification is the topic.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

First conversational pathology copilot. CONCH visual encoder + Llama-2 13B + 456k pathology Q&A instruction-tuning. Cite when discussing instruction-tuned pathology MLLMs.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Owkin's iBOT-pretrained pathology PFM (ViT-B on 6k TCGA WSIs). Open-weights pre-UNI baseline. iBOT recipe = DINO + masked image modelling.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

The first widely-cited pathology vision-language foundation model paper. Releases the OpenPath corpus + open weights. Cite as the canonical pathology VL reference.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Most-cited pre-PFM-era multimodal pathology AI paper. Pan-cancer histology-genomics survival framework: CLAM-style MIL + RNA-seq/mutation/CNV fusion + Cox head across 14 TCGA cancers.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Paige.AI's multimodal generative slide-level PFM. Virchow patch encoder + slide-level Transformer + generative text decoder. Sister to TITAN with explicit slide-level report generation.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Transcriptomics-guided slide-level PFM. CLIP-style contrastive alignment between WSI and bulk RNA-seq. Sibling to report-guided multimodal slide PFMs (TITAN, PRISM).

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Extract technical-focus fields from an article that primarily proposes or evaluates an algorithm, model, training scheme, dataset, or benchmark.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

First widely-cited multimodal whole-slide foundation model. CONCH/UNI patch encoder + slide-level Transformer + report alignment + synthetic captions. Cite when slide-text retrieval / slide captioning / zero-shot slide-level classification are the topic.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

The standard reference for Transformer-based MIL in pathology. Cite when patch-patch correlations are explicitly modelled.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Provide Claude with the rules for citing and using the UNI paper.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Largest published patch-level pathology PFM at release: ViT-H/14, DINOv2, 1.5M MSKCC WSIs. Cite for PFM scaling claims and rare-cancer detection.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

ICIAR 2018 breast-histology classification challenge — 400 microscopy ROIs + 10 WSIs in 4 morphology classes (normal / benign / in-situ / invasive). Small-data stress test for stain robustness and few-shot transfer.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

The seed public benchmark for breast-cancer lymph-node metastasis detection on H&E WSIs (400 slides, two centres). Anchors most weakly-supervised MIL papers.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

NIH/NCI proteomics-focused sister cohort to TCGA. 4.5k FFPE WSIs, 2k patients, 10 cancer cohorts. The canonical external validation cohort for TCGA-trained pathology models, and the only large public cohort with paired H&E + bulk mass-spec proteomics.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Canonical multi-domain mitotic-figure detection benchmark on H&E. Multiple scanners, institutions, tumour types, and species (human + canine). Purpose-built for domain generalisation.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

The standard public benchmark for prostate Gleason / ISUP grading on H&E biopsy WSIs.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Standard public benchmark for nuclei instance segmentation and 5-class classification on H&E patches across 19 tissue types. PQ is the canonical metric.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

The largest public multi-omics cancer cohort: 30k WSIs, 33 cancer types, 11k patients, paired with RNA-seq / mutations / CNV / methylation / clinical records. Default training cohort for almost every pathology AI paper.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Maintain knowledge/graph/{nodes.yaml, edges.yaml} and produce graph.json for the website. Source of truth for graph topology (CLAUDE.md §8).

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Update knowledge/taxonomies/.yaml when a new object introduces a method / task / disease / fusion-pair / clinical category not yet covered.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Default attention-MIL aggregator. Permutation-invariant learned attention pooling over instance embeddings; the foundational pathology MIL aggregator.

Bokai-Zhao/Pathology-Wiki

Skill Claude CodeCodex

Modern ViT-based nuclei instance segmentation + classification. Keeps HoVer-Net's three-decoder head; replaces the encoder with a SAM-pretrained ViT.