{"screening_id":"IVY-CBC-20260619-CA2881","timestamp":"2026-06-19T14:15:21.351803","sex":"M","age":55,"models_run":4,"derived_ratios":{"nlr":1.75,"plr":117.8,"sii":430.0,"siri":0.84},"results":[{"cancer_type":"prostate","display_name":"Prostate Cancer","risk_tier":"Average","probability":0.3202},{"cancer_type":"lung","display_name":"Lung Cancer","risk_tier":"Average","probability":0.3705},{"cancer_type":"crc","display_name":"Colorectal Cancer","risk_tier":"Average","probability":0.3876},{"cancer_type":"multi","display_name":"Ivy Health Alert","risk_tier":"Normal","probability":0.5272}]}
curl --location --request POST 'https://zylalabs.com/api/12952/innovative+ai+cancer+risk+api/25861/cancer+risk+screen' --header 'Authorization: Bearer YOUR_API_KEY'
--data-raw '{
"sex": "M",
"age": 55,
"wbc": 6.2,
"neutrophils": 3650,
"lymphocytes": 2080,
"monocytes": 480,
"rbc": 4.50,
"hemoglobin": 14.1,
"mcv": 89.0,
"rdw": 13.3,
"mchc": 33.5,
"platelets": 245
}'
注册后,每个开发者都会被分配一个个人 API 访问密钥,这是一个唯一的字母和数字组合,用于访问我们的 API 端点。要使用 Innovative AI Cancer Risk API 进行身份验证,只需在 Authorization 标头中包含您的 bearer token。
| 标头 | 描述 |
|---|---|
授权
|
必需
应为 Bearer access_key. 订阅后,请查看上方的"您的 API 访问密钥"。
|
无长期承诺。随时升级、降级或取消。 免费试用包括最多 50 个请求。
(年度计费可节省 2 个月 🎉)
Got it. Here's the clean version — purely informational:
Screen for 20+ cancers from a routine blood test. No imaging. No specialist. No additional tests.
The Ivy CBC Cancer Screen API analyses a standard Complete Blood Count (CBC) and returns AI-powered cancer risk scores across 20+ cancer types in under 50ms. Built for health insurers, corporate wellness platforms, telehealth apps, and lab systems that already collect CBC data — and want to turn it into actionable cancer risk intelligence at scale.
Send 12 CBC parameters (sex, age, and 10 blood values). Receive structured risk tiers and probabilities for four cancer screening models instantly.
Four models run per request:
No PII required. No patient data stored. Results in under 50ms.
The Ivy Health Alert is the API's broadest model. It detects generalised haematological stress patterns associated with a wide basket of malignancies — including many cancers that have no routine early-detection test in standard clinical workflows:
| Category | Cancer Types |
|---|---|
| Haematological | Leukaemia (AML/CLL/ALL), Lymphoma (Hodgkin/NHL), Multiple Myeloma, Myelodysplastic Syndrome, Polycythaemia Vera, Essential Thrombocythaemia |
| Gastrointestinal | Colorectal, Gastric, Pancreatic, Hepatocellular (Liver), Oesophageal |
| Gynaecological | Ovarian, Cervical, Uterine/Endometrial |
| Urological | Prostate, Bladder, Renal Cell (Kidney) |
| Other | Lung, Breast, Thyroid, Sarcomas |
A single blood test. A single API call. 20+ cancers screened simultaneously.
Every model returns one of four tiers based on NHANES population percentiles:
| Tier | Population | Recommended Action |
|---|---|---|
| High | Top 10% | Priority clinical referral |
| Moderate | Next 15% | Ensure screenings are current |
| Average | Middle 25% | Standard preventive care |
| Low | Bottom 50% | Routine monitoring |
| Response Time | < 50ms |
| Availability | 99.9% SLA |
| Hosting | AWS London (eu-west-2) |
| Auth | X-Api-Key header |
| Data Retention | Zero — CBC values discarded after response |
| PII Required | None |
| White-label | Yes — raw JSON, present under your own brand |
癌症风险筛查端点基于标准的全面血液检查(CBC)返回超过20种癌症类型的人工智能驱动的癌症风险评分。它为前列腺或乳腺癌、肺癌、结直肠癌和多癌信号提供风险等级分类(高、中、平均、低)
Key fields in the response include `screening_id`, `timestamp`, `sex`, `age`, `models_run`, `derived_ratios`, and `results`. The `results` field contains detailed risk information for each cancer type assessed.
The response data is structured in JSON format. It includes a top-level object with metadata (like `screening_id` and `timestamp`) and a `results` array that lists each cancer type, its display name, risk tier, and probability.
The endpoint accepts 12 parameters: `sex`, `age`, and 10 CBC blood values (wbc, neutrophils, lymphocytes, monocytes, rbc, hemoglobin, mcv, rdw, mchc, platelets). Each parameter must adhere to specified ranges and formats.
The data is derived from proprietary machine learning models trained on NHANES population data. This ensures that the risk assessments are based on a robust dataset reflecting diverse population health metrics.
Typical use cases include enhancing health insurer screenings, integrating cancer risk assessments into corporate wellness programs, enriching telehealth applications, and providing silent background screening in lab systems for CBC tests.
用户可以通过解读风险等级来利用返回的数据,以指导临床转诊、确保及时筛查或根据风险水平监测患者。这些概率可以为医疗决策和患者管理策略提供信息
Data accuracy is maintained through continuous model training and validation against real-world health outcomes. Regular updates and quality checks ensure that the risk assessments remain reliable and relevant to current health trends.