北仑:台州解析

kernel/src/test/java/com/lantone/qc/kernel/tzAdmissionNote.java

@@ -0,0 +1,386 @@
+package com.lantone.qc.kernel;
+
+import com.google.common.collect.Lists;
+import com.lantone.qc.kernel.structure.ai.model.Lemma;
+import com.lantone.qc.pub.util.StringUtil;
+import com.lantone.qc.trans.comsis.OrdinaryAssistant;
+import com.lantone.qc.trans.util.http.db.DBUtil;
+
+import java.io.BufferedReader;
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.InputStreamReader;
+import java.sql.*;
+import java.util.*;
+import java.util.regex.Matcher;
+import java.util.regex.Pattern;
+
+public class tzAdmissionNote {
+
+    public static void main(String[] args) {
+        String path = "C:/Users/dev_1/Desktop/dcsj/test.txt";
+        String str = readFile(path);
+//        List<Map<String, String>> xmlMaps = WritXmlSelect();
+//        System.out.println(xmlMaps);
+//        AcquirePatientName(xmlMaps);
+    }
+
+    // 读取文件内容
+    public static String readFile(String path) {//路径
+        File file = new File(path);
+        StringBuffer result = new StringBuffer();
+        List<String> caseList = new ArrayList<>();
+        Map<String, String> structureMap = new HashMap<>();
+        try {
+            BufferedReader br = new BufferedReader(new InputStreamReader(new FileInputStream(file), "UTF-8"));//构造一个BufferedReader类来读取文件
+            String line = null;
+            while ((line = br.readLine()) != null) {//使用readLine方法，一次读一行
+                boolean flag = cutCase(line, caseList, result);
+                if (flag) {
+                    result = new StringBuffer();
+                    result.append(line);
+                }
+            }
+            if (caseList.size() > 0) {
+                analysisList(caseList, structureMap);
+            }
+            br.close();
+        } catch (Exception e) {
+            e.printStackTrace();
+        }
+        return result.toString();
+    }
+
+    //切出每一个个病历
+    private static void analysisList(List<String> caseList, Map<String, String> structureMap) {
+        for (String content : caseList) {
+            String[] split = content.split("\t");
+            if (split.length > 4) {
+                structureMap.put("病案号", split[0]);
+                if (split[1].equals("F")) {
+                    structureMap.put("性别", "女");
+                } else {
+                    structureMap.put("性别", "男");
+                }
+                structureMap.put("年龄", split[2] + split[3]);
+                //主诉，现病史等的处理
+                analysis(split[4], structureMap);
+            }
+        }
+    }
+
+    public static void analysis(String content, Map<String, String> structureMap) {
+        List<String> titles = Lists.newArrayList("主诉", "现病史", "既往史", "个人史", "月经史", "婚育史", "家族史", "体格检查",
+                "精神检查", "诊断", "处理", "处理结果", "注意事项", "西药", "检查", "外阴", "阴道", "宫颈", "子宫", "双附件");
+        if (StringUtil.isNotBlank(content)) {
+            List<String> sortTitles = sortTitles(titles, content);
+            cutByTitles(content, sortTitles, 0, structureMap, true);
+            structureMap = OrdinaryAssistant.mapKeyContrast(structureMap, consultationResults_pageDataTitles);
+            storage(structureMap);
+        }
+
+    }
+
+    private static List<String> consultationResults_pageDataTitles = Lists.newArrayList(
+            "处理结果=处理"
+    );
+
+
+    private static void storage(Map<String, String> structureMap) {
+        Connection conn = null;
+        Statement stmt = null;
+        PreparedStatement ps = null;
+        ResultSet rs = null;
+        String sql = "insert into str_admission_note(name,price,bookCount,author) values(?,?,?,?)";
+        try {
+            conn = DBUtil.getConnection();
+            ps = conn.prepareStatement(sql);
+
+//            ps.setString(1, );
+//            ps.setDouble(2, );
+//            ps.setInt(3, );
+//            ps.setString(4, );
+            int row = ps.executeUpdate();
+        } catch (Exception e) {
+            e.printStackTrace();
+        } finally {
+            try {
+                if (rs != null) {
+                    rs.close();
+                    rs = null;
+                }
+                if (stmt != null) {
+                    stmt.close();
+                    stmt = null;
+                }
+                if (conn != null) {
+                    conn.close();
+                    conn = null;
+                }
+            } catch (SQLException e) {
+                e.printStackTrace();
+            }
+        }
+    }
+
+
+    /**
+     * 将title根据在文本中的位置排序
+     *
+     * @param titles
+     * @param content
+     * @return
+     */
+    public static List<String> sortTitles(List<String> titles, String content) {
+        Map<Integer, String> titleIndex = new TreeMap<>();
+        int index, index_1, index_2;
+        for (String title : titles) {
+            index_1 = content.indexOf(title + "：");
+            index_2 = content.indexOf(title + ":");
+            index = Math.max(index_1, index_2);
+            if (index != -1) {
+                titleIndex.put(index, title);
+                StringBuffer sb = new StringBuffer(title.length());
+                for (int i = 0; i < title.length(); i++) {
+                    sb.append('*');
+                }
+                content = content.substring(0, index) + sb.toString() + content.substring(index + title.length() + 1);
+                //                content = content.substring(0, index) + content.substring(index + title.length() + 1);
+            }
+        }
+        titles = Lists.newArrayList(titleIndex.values());
+        return titles;
+    }
+
+    /**
+     * 根据文书各标题截取相应文本，存入structmap中
+     *
+     * @param line         原始文本
+     * @param titles       文书各标题
+     * @param depth        递归深度，也就是titles取值时的下标值
+     * @param structureMap 存储结构化数据
+     */
+    public static void cutByTitles(String line, List<String> titles, int depth, Map<String, String> structureMap, Boolean flag) {
+        if (depth > titles.size() || titles.size() == 0) {
+            return;
+        }
+        String beforeTitle = null, title = null, newTitle = null, value = null;
+        beforeTitle = titles.get(Math.max(depth - 1, 0));
+        if (line.indexOf(beforeTitle) != 0 && flag) {
+            structureMap.put("现病史", line.substring(0, line.indexOf(beforeTitle)));
+        }
+        title = titles.get(Math.min(depth, titles.size() - 1));
+        if (depth == titles.size()) {
+            /*if (line.contains("\n")) {
+                line = line.split("\n")[0];
+            }
+            */
+            value = line.replace("\n", "");
+            structureMap.put(beforeTitle, value.trim());
+            return;
+        }
+        if (line.contains(title + "：") || line.contains(title + ":")) {
+            if (line.contains(title + "：")) {
+                newTitle = title + "：";
+            } else {
+                newTitle = title + ":";
+            }
+            if (depth > 0) {
+                value = line.substring(0, line.indexOf(newTitle));
+                structureMap.put(beforeTitle.replace(" ", ""), value.trim().replace("\n", ""));
+            }
+            line = line.substring(line.indexOf(newTitle) + newTitle.length());
+            depth++;
+        } else {
+            titles.remove(depth);
+        }
+        cutByTitles(line, titles, depth, structureMap, false);
+    }
+
+
+    //切出每一个个病历
+    private static boolean cutCase(String line, List<String> caseList, StringBuffer result) {
+        line = line.replace("浙江省台州市立医院", "")
+                .replace("台州学院医学院附属市立医院", "")
+                .replace("-------------------------------------------------------------------------------------------", "");
+        if (line.contains("\t")) {
+            String content = line.substring(0, line.indexOf("\t"));
+            String regex = "^\\d{7,}$";
+            if (content.matches(regex)) {
+                caseList.add(result.toString());
+                return true;
+            }
+        }
+        result.append(line);
+        return false;
+    }
+
+    //取患者姓名
+    private static void AcquirePatientName(List<Map<String, String>> xmlMaps) {
+        for (Map<String, String> structMap : xmlMaps) {
+            String recId = structMap.get("Id");
+            String xmlText = structMap.get("xml文书");
+            if (StringUtil.isNotBlank(xmlText)) {
+                //患者姓名
+                String patientName = null;
+                if (xmlText.contains("</NewCtrl>")) {
+                    String[] newCtrls = xmlText.split("<NewCtrl");
+                    for (int i = 1; i < newCtrls.length; i++) {
+                        if (newCtrls[i].contains("患者姓名")) {
+                            int conTextStartInt = newCtrls[i].indexOf("<Content_Text>") + 14;
+                            int conTextEndInt = newCtrls[i].indexOf("</Content_Text>");
+                            //取得患者姓名
+                            patientName = newCtrls[i].substring(conTextStartInt, conTextEndInt);
+                            break;
+                        }
+                    }
+                } else if (xmlText.contains("</EMR-TERM>")) {
+                    String[] emrTerms = xmlText.split("<EMR-TERM");
+                    for (int i = 1; i < emrTerms.length; i++) {
+                        if (!emrTerms[i].contains("任意医生")) {
+                            int conTextStartInt = emrTerms[i].indexOf("creator=\"") + 9;
+                            int conTextEndInt = emrTerms[i].indexOf("createDate") - 2;
+                            //取得患者姓名
+                            patientName = emrTerms[i].substring(conTextStartInt, conTextEndInt);
+                            break;
+                        }
+                    }
+                }
+                //名字脱敏
+                if (StringUtil.isNotBlank(patientName)) {
+                    xmlText = ReplacePatientName(xmlText, patientName);
+                    //手机号匹配
+                    Pattern p = Pattern.compile("1[345678]\\d{9}");
+                    Matcher m = p.matcher(xmlText);
+                    if (m.find()) {
+                        String group = m.group();
+                        char[] groupChars = group.toCharArray();
+                        String groupPhone = String.valueOf(groupChars[0]) + String.valueOf(groupChars[1]) + String.valueOf(groupChars[2]) + "********";
+                        xmlText = xmlText.replace(group, groupPhone);
+                    }
+                    //座机号匹配
+                    Pattern p2 = Pattern.compile("0\\d{2,3}-[1-9]\\d{6,7}");
+                    Matcher m2 = p2.matcher(xmlText);
+                    if (m2.find()) {
+                        String group = m2.group();
+                        String subStr = xmlText.substring(xmlText.indexOf(group) - 100, xmlText.indexOf(group));
+                        if (subStr.contains("家庭电话")) {
+                            char[] groupChars = group.toCharArray();
+                            String groupPhone = null;
+                            if (groupChars.length == 12) {
+                                groupPhone = String.valueOf(groupChars[5]) + String.valueOf(groupChars[6]) + String.valueOf(groupChars[7]) + "****";
+                            } else if (groupChars.length == 13) {
+                                groupPhone = String.valueOf(groupChars[5]) + String.valueOf(groupChars[6]) + String.valueOf(groupChars[7]) + "*****";
+                            }
+                            xmlText = xmlText.replace(group, groupPhone);
+                        }
+                    }
+                    //身份证匹配
+                    Pattern p1 = Pattern.compile("\\d{17}[\\d|x]|\\d{15}");
+                    Matcher m1 = p1.matcher(xmlText);
+                    if (m1.find()) {
+                        String idNumber = m1.group();
+                        String subStr = xmlText.substring(xmlText.indexOf(idNumber) - 5, xmlText.indexOf(idNumber));
+                        if (!subStr.contains("Id=")) {
+                            char[] idNumberChars = idNumber.toCharArray();
+                            String idNumberSen = String.valueOf(idNumberChars[0]) + String.valueOf(idNumberChars[1]) + String.valueOf(idNumberChars[2]) + String.valueOf(idNumberChars[3]) + "**************";
+                            xmlText = xmlText.replace(idNumber, idNumberSen);
+                        }
+                    }
+                    System.out.println(recId);
+                    System.out.println(xmlText);
+                    ReplaceSensitivity(recId, xmlText);
+                }
+            }
+        }
+    }
+
+    private static void ReplaceSensitivity(String recId, String xmlText) {
+        Connection conn = null;
+        Statement stmt = null;
+        ResultSet rs = null;
+        try {
+            conn = DBUtil.getConnection();
+            stmt = conn.createStatement();
+            stmt.executeUpdate("UPDATE med_medical_record_content mmrc SET mmrc.xml_text='" + xmlText + "'" + "WHERE mmrc.rec_id='" + recId + "'");
+        } catch (Exception e) {
+            e.printStackTrace();
+        } finally {
+            try {
+                if (rs != null) {
+                    rs.close();
+                    rs = null;
+                }
+                if (stmt != null) {
+                    stmt.close();
+                    stmt = null;
+                }
+                if (conn != null) {
+                    conn.close();
+                    conn = null;
+                }
+            } catch (SQLException e) {
+                e.printStackTrace();
+            }
+        }
+    }
+
+    //查询xml文件内容
+    public static List<Map<String, String>> WritXmlSelect() {
+        Connection conn = null;
+        Statement stmt = null;
+        ResultSet rs = null;
+        List<Map<String, String>> resultList = new ArrayList<>();
+        try {
+            conn = DBUtil.getConnection();
+            stmt = conn.createStatement();
+            rs = stmt.executeQuery(
+                    "SELECT mmrc.rec_id,mmrc.xml_text FROM `med_medical_record_content` mmrc\n" +
+                            "LEFT JOIN med_medical_record mmr ON mmr.mode_id not in(\n" +
+                            "'16','20','21','52','53','54','36','56') WHERE mmrc.rec_id = mmr.rec_id");
+            while (rs.next()) {
+                Map<String, String> map = new LinkedHashMap<>();
+                map.put("Id", rs.getString(1));
+                map.put("xml文书", rs.getString(2));
+                resultList.add(map);
+            }
+        } catch (Exception e) {
+            e.printStackTrace();
+        } finally {
+            try {
+                if (rs != null) {
+                    rs.close();
+                    rs = null;
+                }
+                if (stmt != null) {
+                    stmt.close();
+                    stmt = null;
+                }
+                if (conn != null) {
+                    conn.close();
+                    conn = null;
+                }
+            } catch (SQLException e) {
+                e.printStackTrace();
+            }
+        }
+        return resultList;
+    }
+
+    //名字脱敏
+    private static String ReplacePatientName(String xmlText, String patientName) {
+        char[] patientNameChars = patientName.toCharArray();
+        String sensitivityName = null;
+        if (patientNameChars.length == 2) {
+            sensitivityName = patientNameChars[0] + "*";
+        } else if (patientNameChars.length == 3) {
+            sensitivityName = patientNameChars[0] + "**";
+        } else if (patientNameChars.length == 4) {
+            sensitivityName = patientNameChars[0] + "***";
+        }
+        if (StringUtil.isNotBlank(sensitivityName)) {
+            xmlText = xmlText.replace(patientName, sensitivityName);
+        }
+        return xmlText;
+    }
+}

trans/src/main/java/com/lantone/qc/trans/util/http/db/DBUtil.java

@@ -21,7 +21,7 @@ public class DBUtil {
             //1.加载驱动程序
             Class.forName("com.mysql.jdbc.Driver");
             //2.获得数据库的连接
-            connection = DriverManager.getConnection("jdbc:mysql://192.168.2.122:3306/qc?useUnicode=true&characterEncoding=utf8&useSSL=false", "root", "lantone");
+            connection = DriverManager.getConnection("jdbc:mysql://192.168.2.236:3306/tz_admission_note?useUnicode=true&characterEncoding=utf8&useSSL=false", "root", "lantone");
         } catch (ClassNotFoundException e) {
             e.printStackTrace();
         } catch (SQLException e) {