深入解析：MCoT在医疗AI工程化编程的实践手册（中）

2.4 核心代码实现

A. 定义事件（使用Avro/Protobuf Schema）

// inference_events.proto
syntax = "proto3";
message InferenceRequest {
  string request_id = 1;
  string patient_id_hash = 2;
  repeated DataSource data_sources = 3;
  string workflow_definition_id = 4; // e.g., "dr_diagnosis_v1"
}
message TaskCompleted {
  string task_id = 1;
  string request_id = 2;
  string worker_id = 3;
  string stage_name = 4; // "localization"
  string artifact_uri = 5; // s3://evidence-bucket/loc/123.json
  bool success = 6;
  string error_message = 7;
  string model_version = 8;
}
message ValidationResult {
  string task_id = 1;
  string request_id = 2;
  bool passed = 3;
  map metrics = 4; // e.g., {"dice_score": 0.82}
  string decision = 5; // "PROCEED", "FALLBACK", "RETRY"
}

B. Worker示例（C++ + ONNX Runtime）

// localization_worker.cpp
#include <onnxruntime_cxx_api.h>
  #include <opencv2/opencv.hpp>
    #include <kafkacpp/kafka_producer.h> // A hypothetical Kafka C++ library
      void run_localization_task(const std::string& task_json) {
      
      // 1. Parse task, download image from S3/MinIO
      Task task = parse_task(task_json);
      cv::Mat image = download_image(task.get_input_image_uri());
      // 2. Preprocess
      cv::Mat resized_img;
      cv::resize(image, resized_img, cv::Size(224, 224));
      // ... other preprocessing steps
      // 3. Run ONNX Model
      Ort::Env env(ORT_LOGGING_LEVEL_WARNING, "LocalizationWorker");
      Ort::Session session(env, L"localization_model_v1.onnx", Ort::SessionOptions{
      });
      // ... create input tensor, run session, get output tensor ...
      // 4. Postprocess results (e.g., NMS, convert to bbox)
      LocalizationResult result = postprocess_bboxes(output_tensor);
      // 5. Save intermediate artifact
      std::string artifact_uri = save_artifact_to_s3(result.to_json(), task.request_id);
      // 6. Publish TaskCompleted event
      TaskCompletedEvent event;
      event.set_task_id(task.id);
      event.set_request_id(task.request_id);
      event.set_stage_name("localization");
      event.set_artifact_uri(artifact_uri);
      event.set_success(true);
      event.set_model_version("localization_model_v1.onnx");
      KafkaProducer producer("task_completed_topic");
      producer.produce(event.serialize());
      }

C. 验证网关

# validation_gateway.py
from scipy.spatial.distance import dice
import numpy as np
def validate_localization(task_completed_event: TaskCompleted):
"""
Validates a localization task by comparing against a ground truth
if available, or using other heuristics.
"""
ground_truth_uri = get_ground_truth_uri(task_completed_event.request_id)