Oncoschema
Schema package for oncology extraction from cancer clinical documents.
Structure
📁 oncoschema
├── examples/ # Training examples showing document input and structured output
├── schema.py # Pydantic model for specifying expected output structure
├── prompt_builder.py # Prompt builder for data generation and inference
├── prompt_datagen.txt # Prompt template with example (for training data generation)
├── prompt_main.txt # Prompt template without example (for inference/deployment)
└── py.typed # Type checking marker
Usage
from oncoschema.prompt_builder import PromptBuilder
# Initialize builder
builder = PromptBuilder()
# Build data generation prompt (with example)
datagen_prompt = builder.build_datagen_prompt()
# Build main/inference prompt (without example)
main_prompt = builder.build_main_prompt()
Schema
| Type | Values |
|---|---|
| TopographyType | unknown_primary, other, haematological, lung, pleura, other_respiratory, oesophagus, stomach, small_intestine, colon, rectum, pancreas, liver, gallbladder, bile_duct, other_gi, kidney, bladder, prostate, testis, other_gu, breast, cervix, uterus, ovary, other_gynae, brain, spinal_cord, other_cns, oral, hypo_oro_naso_pharynx, larynx, salivary_gland, nasal_cavity, paranasal_sinus, thyroid, adrenal_gland, other_endocrine, skin, soft_tissue, bone |
| MorphologyType | unknown_morphology, other, adenocarcinoma, squamous_cell_carcinoma, urothelial_carcinoma, renal_cell_carcinoma, hepatocellular_carcinoma, small_cell_carcinoma, non_small_cell_carcinoma, carcinoma_other, mesothelioma, melanoma, neuroendocrine, carcinoid, sarcoma_nos, gastrointestinal_stromal_tumour, osteosarcoma, chondrosarcoma, ewing_sarcoma, rhabdomyosarcoma, kaposi_sarcoma, soft_tissue_sarcoma, acute_lymphoblastic_leukaemia, acute_myeloid_leukaemia, chronic_lymphocytic_leukaemia, chronic_myeloid_leukaemia, hodgkin_lymphoma, non_hodgkin_lymphoma, multiple_myeloma, leukaemia_other, myelodysplastic, glioblastoma, astrocytoma, oligodendroglioma, meningioma, seminoma, teratoma, choriocarcinoma, wilms_tumour, retinoblastoma, hepatoblastoma |
| MSIStatus | msi_high, ms_stable |
| TMBStatus | tmb_high, tmb_low, tmb_intermediate |
| MolecularBiomarkerType | other, braf, ntrk1, ntrk2, ntrk3, ret, erbb2_her2, tp53, brca1, brca2, mlh1, msh2, msh6, pms2, palb2, rad51, egfr, alk, ros1, met, kras, nras, pik3ca, esr1_er, pgr_pr, ki_67, kit, pdgfra, fgfr1, fgfr2, fgfr3, idh1, idh2, pdl1, nf1, nf2, mgmt, npm1, flt3, bcr_abl1, jak2, bcl2, myc |
| BiomarkerStatus | altered, negative, equivocal, hypothetical |
| CancerScoreName | pathological_grade, gleason, figo, dukes, breslow, clark, binet, child_pugh, other |
| SpreadType | other, lymph_node, liver, lung, spine, other_bone, brain, other_cns, adrenal, kidney, pleura, peritoneum, skin, pancreas, spleen, ovary, testis, thyroid, stomach, bowel, bladder, prostate, breast, head_and_neck |
| TimelineEventType | had_systemic_or_radiotherapy_treatment, had_surgical_treatment_performed, experienced_toxicity_or_complication_related_to_treatment, experienced_treatment_reduction_or_stop, considered_for_clinical_trial, enrolled_to_clinical_trial, positive_treatment_response_on_assessment, evidence_of_metastatic_progression, radiology_evidence_of_disease_progression, experienced_disease_remission, experienced_disease_recurrence, patient_died |
| PatientFindingType | comorbidity_finding, social_or_family_finding, symptom_finding, physical_examination_finding, functional_finding, mental_state_finding |
| PatientFindingStatus | is_present, is_not_present, uncertain |
| FuturePlanType | planned_systemic_or_radiotherapy_treatment, planned_surgery_treatment, planned_investigation, planned_clinical_trial_involvement |
License
This project uses a proprietary license issued by Guy's and St Thomas' NHS Foundation Trust, enabling free (non-commercial) use by NHS organisations. See LICENSE files for details.
Metadata
Release files for londonaicentre-oncoschema 2.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| londonaicentre_oncoschema-2.1.0.tar.gz | 24.4 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| londonaicentre_oncoschema-2.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 50.9 kB
Release files / londonaicentre_oncoschema-2.1.0.tar.gz
| Download URL | londonaicentre_oncoschema-2.1.0.tar.gz |
|---|---|
| Size | 24.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
fddd5c23430bfb618eff86b099750b01b4847e47e26018aaf6f0e0b02426ad0f
|
|
BLAKE2b-256 checksum How to use checksums |
ed480121db9f51e791354fad7031336236ec70c3f5dbaa8449fe5974b214a2c8
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
uv/0.11.21 {"installer":{"name":"uv","version":"0.11.21","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Amazon Linux","version":"2023","id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
Release files / londonaicentre_oncoschema-2.1.0-py3-none-any.whl
| Download URL | londonaicentre_oncoschema-2.1.0-py3-none-any.whl |
|---|---|
| Size | 26.5 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
16495ff5f2506f8cd6008d99b2a4cd4e1e77976ede93f3b71fddb9001d31073b
|
|
BLAKE2b-256 checksum How to use checksums |
71472617cdf8f5e78e3a724ea4b051f6f8558e2bd028fd9d6c30e4d223808362
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
uv/0.11.21 {"installer":{"name":"uv","version":"0.11.21","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Amazon Linux","version":"2023","id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
|
