摘要
如今随着科技、经济水平的不断发展，人们对健康的需求指数也越来越高。但目前，病患主要的问诊形式依旧是以前往目标医院进行咨询的传统模式为主。这种形式将极大的浪费患者的时间，同时，也正是这种医疗模式使得很多人小病不医致使最终酿成大病。目前普通医疗体系不能永远保持精确又快速的诊断,通过机器学习辅助医疗不仅能大幅削减成本, 其诊断结果几乎能实时获取。越来越多的情况下，通过机器学习进行的诊断能够比老练的医生提供更准确的诊断。开发一款基于机器学习的医疗APP，初衷在于解决看病难，看病浪费时间大的问题。无论在何地，通过APP就能进行医疗诊断，为患者提供相应的治疗方案、注意事项与健康生活习惯计划，为用户的小病进行定制化治疗，帮助用户早日恢复健康。若诊断判定病症较严重，将提醒患者尽早到医院就医，以保证安全性。
本文采用基于TensorFlow的机器学习模型实现在线诊断，并且向用户提供完整的医疗服务链，包括预约、诊断、咨询、购药等功能。系统可以分为App端和Bmob管理后台，App端使用Android原生的Bottom Navigation Activity模板进行编写，后台采用Bmob云数据库对系统各表数据和用户权限进行管理。
本系统实现了用户登录注册、管理预约就诊信息、购买药品及查看订单、医患交流、医院预约挂号、医院导航、浏览健康资讯、查询疾病、医院、药品等信息以及在线诊断等功能，大大提高了患者就诊效率，达到了系统设计的预期。
关键词：互联网医疗  机器学习  自主诊断  服务链
Abstract
Nowadays, with the continuous development of science and technology and economic level, people's demand index for health is getting higher and higher. But at present, the main form of patient consultation is still based on the traditional mode of consulting at the target hospital. This form will greatly waste the patient's time. At the same time, it is this medical model that makes many people suffer from minor illnesses and eventually become serious illnesses. At present, the general medical system cannot always maintain accurate and rapid diagnosis. The aid of medical learning through machine learning can not only significantly reduce costs, but also obtain the diagnosis results in real time.In more and more cases, diagnosis through machine learning can provide a more accurate diagnosis than experienced doctors. Develop a medical APP based on machine learning, the original intention is to solve the problem of difficulty in seeing a doctor and wasting time in seeing a doctor. No matter where you are, you can make medical diagnosis through APP, provide patients with corresponding treatment plans, precautions and healthy lifestyle plans, and customize treatment for users' minor illnesses to help users recover health as soon as possible. If the diagnosis determines that the condition is serious, the patient will be reminded to go to the hospital as soon as possible to ensure safety.
This article uses a TensorFlow-based machine learning model to achieve online diagnosis and provide users with a complete medical service chain, including appointment, diagnosis, consultation, drug purchase and other functions. The system can be divided into App side and Bmob management background. App side uses Android's native Bottom Navigation Activity template to write. The background uses Bmob cloud database to manage the system table data and user permissions.
This system realizes functions such as user login and registration, management of appointment information, purchase of medicines and order viewing, doctor-patient communication, hospital appointment registration, hospital navigation, browsing health information, querying diseases, hospitals, drugs and other information, and online diagnosis This improves the efficiency of patient visits and meets the expectations of system design.
Key Words：Internet Medical  Machine Learning  Self-diagnosis  Service Chain
目 录
摘要 I
Abstract II
目 录 III
图清单 V
表清单 VI
1 绪论 1
1.1 课题意义和目标 1
1.2 国内外研究现状 1
1.3 论文的主要工作 2
1.4 论文的组织结构 2
1.5 本章小结 3
2 系统分析 4
2.1 可行性分析 4
2.2 需求分析 5
2.3 方案比选 10
2.3.1 方案一 10
2.3.2 方案二 11
2.4 本章小结 11
3 系统的设计 12
3.1 软件体系结构 12
3.2 功能设计 12
3.3 持久化设计 14
3.4 社会健康、文化、法律相关设计 17
3.5 本章小结 18
4 系统的实现 19
4.1 医疗诊断 19
4.2 用户操作管理 22
4.3 医院服务 24
4.4 医疗信息显示和查询 26
4.5 本章小结 28
5 系统运行与效果分析 29
5.1 界面设计概要 29
5.2 用户管理 29
5.3 诊断购药 30
5.4 医院医生服务 33
5.5 医疗信息管理 35
5.6 Bmob管理后台 37
5.7 本章小结 37
6 系统测试 37
6.1 测试方法 38
6.2 测试方案及计划 38
6.3 测试过程及结果分析 39
6.4 本章小结 39
7 总结与展望 41
7.1 总结 41
7.2 展望 41
参考文献 42
致谢 43