영남의대학술지제24권제2호 Yeungnam Univ. J. of Med. Vol.24 No.2 p137-147, Dec. 2007 원저 건강검진자를대상으로해석적보고를위한전문가시스템의개발 이채훈영남대학교의과대학진단검사의학교실 Development of Rule-based Expert System for Interpretative Report with Health Screening Tests Chae Hoon Lee Department of Laboratory Medicine, College of Medicine, Yeungnam University, Daegu, Korea -Abstract- Background:Interpretative reporting is an important aspect of laboratory medicine. The large menu of laboratory tests available today makes it increasingly difficult for the non-specialist to order and interpret all laboratory tests. The aim of this study was to determine the usefulness of an expert system to interpret laboratory tests and help physicians order the appropriate tests. Materials and Methods:In order to interpret laboratory tests, a rules-based expert system was developed. In this module, if-then rules were used to interpret the given test result patterns (e.g. urinalysis, anemia, hepatitis B virus, hypercholesterolemia, glucose, syphilis, and tumor markers) and select matching text elements. The system was used to evaluate 535 subjects who visited a health-check program. 1) Results:The overall abnormal rate was 50.5% in the expert system; 34% for cholesterol, 9.9% for urinalysis, 8.0% for anemia, 7.7% for thyroid function tests, 4.5% for tumor marker study, 4.7% for hepatitis virus antigen, 4.3% for serum glucose, and 1.1% for syphilis. Conclusion:These results indicate that the application of the expert system for the 책임저자 : 이채훈, 대구시남구대명5동 317-1, 영남대학교의과대학진단검사의학교실 Tel: (053) 620-3632, Fax: (053) 653-7774, E-mail: chlee@med.yu.ac.kr 137
- 이채훈 - interpretation of laboratory tests may provide a useful method for the interpretation of reports. However more rules are needed for the application to in-patients. Key Words: Expert system, Interpretation, Knowledge-based system. 서론검사결과의해석은진단검사의학과의필수적인업무중의하나이다. 1) 하지만일일발생하는검사건수가많아대부분의병원에서모든검사에대한해석적보고를실시하지는못하고, 혈액도말, 골수판독, 전기영동등부분적인해석적보고를실시하고있는실정이지만최근국내외에서해석적보고의영역을점차적으로확대하려는노력이시도되고있다. 2-6) 이러한해석적보고의시도는새로운검사의증가와복잡성으로임상의가진료에필요한새로운검사지식을신속하게습득하여환자진 료에적절하게이용하는것이때때로어렵고, 특히응급실이나중환자실과같은응급을요하는중환자가많은상황에서중요한검사결과를간과하거나미쳐분석하지못하여오진을범하는경우가많기때문에필요성이대두되고있다. 7) 또한향후새로운검사나복잡한검사가도입되는경우더욱더해석적보고에대한요구가증가될것으로예상되고있다. 8) 검사결과의해석은궁극적으로환자개개인에맞는형태가되어야하며이경우충분한환자의임상적정보를알고있어야하는것이 1, 9, 10) 가장중요하지만, 시간과공간적으로어려움이있다. 따라서일차적해석적보고는환 Fig. 1. Expert rules setting screen showing comment lists according to test result pattern. 138
- 건강검진자를대상으로해석적보고를위한전문가시스템의개발 - Table 1. Example of knowledge-based expert system, thyroid function tests If clause var RESULT_VALUE:string; begin RESULT_VALUE:=''; if([s_tsh]>4.0)and (([S_FreeT4]>0.9)and([S_FreeT4]<2.0))and (([S_FreeT3]>1.5)and([S_FreeT3]<5.0))then begin RESULT_VALUE:='A'; end else if([s_tsh]>4.0)and([s_freet4]<0.9)then begin RESULT_VALUE:='AB'; end else if([s_tsh]<0.23)then begin if ([S_FreeT4]>2.0)then begin RESULT_VALUE:='B'; end else if(([s_freet4]>0.9)and([s_freet4]<2.0))then begin RESULT_VALUE:='BB'; end; end else begin if ([S_FreeT4]>2.0)then begin RESULT_VALUE:='CA'; end; end; result:=result_value; end; Action list according to RESULT_VALUE A AB B BB C Euthyroid Sick Syndrome, subclinical hypothyroidism, transient Thyroid 질환이의심되면추가적으로 Thyroid Antibody(Thyroglobin) 의검사로 Autoimmune Thyroiditis의감별이필요합니다. R/O) Primary hypothroidism, Thyroid Ablation Removal, Euthyroid Sick Syndrome, Transient thyroiditis Hyperthyroidism. rec) Thyroid stimulating Immunoglobulin(TSI), Radioiodine Thyroid Scan and % uptake. Euthyroid Sick Syndrome, subclinical hypothyroidism, transient Thyroid 질환이의심되면추가적으로 Thyroid Antibody(Thyroglobin) 의검사로 Autoimmune Thyroiditis의감별이필요합니다. follow up) TSH, Free T4: 6 month later Abbreviations: S_TSH, serum thyroid stimulating hormone; S_FreeT4, serum free thyroxine; S_FreeT3, serum free triiodothyronine. 자의임상적정보없이 Expert system( 전문가시스템 ) 을이용하여보고하는경향이있으며, 9-11) 국내에서도부분적으로전문가시스템을이용 2, 5, 6) 하여해석을시도하고있지만, 모든검사에대한해석적보고는실시하고있지못하고있는실정이다. 이에본연구는 rule-based expert system을 개발하여주로묶음검사형태로되어있는종합건진검사들을대상으로일차적해석적보고의유용성을살펴보고향후입원및외래방문환자의검사판독에도입하여보다유용한검사결과형태를보고함으로써환자의진단과치료에도움이되고자하였다. 139
- 이채훈 - Table 2. Examples of final interpretative comments Example 1. <HBsAg(-), Anti-HBsAB(+), Anti-HBc(+)> 과거 B 형간염바이러스의감염이있었으며, 현재 B 형간염바이러스에대해면역을지니고있습니다. <TSH: 9.25, Free T4:1.2, Free T3: 4.08> Euthyroid Sick Syndrome, subclinical hypothyroidism, transient Thyroid 질환이의심되면추가적으로 Thyroid Antibody(Thyroglobin) 의검사로 Autoimmune Thyroiditis 의감별이필요합니다. <GGT:98, Glucose: 114, TG:237> Triglyceride 는일차적으로고지혈증에증가되어나타나지만, 타질환 ( 비만, 조절이되지않은당뇨, 알콜등 ) 에의해이차적인증가될수있으며상기환자의경우당뇨및알콜로인한증가로사료됩니다 Example 2. <HBsAg(-), Anti-HBsAB(+), Anti-HBc(+)> 과거 B 형간염바이러스의감염이있었으며, 현재 B 형간염바이러스에대해면역을지니고있습니다. <Glucose: 136> 혈당측정시최소한 8 시간의금식후측정해야정확한결과를얻을수있으며, Oral GTT 와 C-peptide 검사를추가적으로시행하여야합니다. <HbA1c> Hemoglobin A1C 는과거 2-3 개월전의당뇨조절여부를보여주는검사로당뇨병이있는경우 7% 이하를유지하는것이좋습니다.7% 이하인경우연 2 차례, 7% 이상인경우최소한 3 개월마다검사를시행하는것이당조절에도움이됩니다. <CEA:40.5> CEA 는 gastrointestinal (GI), 특히 colorectal cancer 에증가되는경우가많으며, 그외 lung, breast, liver, pancreas, stomach, and ovary cancer 에서도증가될수있으므로추가적검사가필요합니다. 대상및방법 1. 검사판독을위한룰의구축검사판독을위한 algorithm은검사에따른감별질환, 검사의유용성에관해보고한문헌 들을 12-15) 참조하여 rule-based expert system 으로구축하였다. 전문가룰은요검사, 혈액검사 ( 빈혈관련, 기타혈액질환 ), 일반생화학검사 ( 간기능, 신장기능등 ), 바이러스감염검사 (B 형간염바이러스, C형간염바이러스등 ), 일 반적감염지표 ( 백혈구수, C-reactive protein), 종양표지자및갑상선기능검사등의 42개룰로구성하였으며, 각각의룰은 1개에서 6개까지의반환값을갖게하였다. 룰의구성을조건부분과실행부분으로구분하였으며, 조건부분은검체및각검사항목의결과에따라분류하였으며, 실행부분은조건부분에서받은반환값에따라감별질환, 주석및추가검사가보고시출력되도록하였다 (Fig. 1, Table 1). 사용자인테페이스는병록번호선택시병원전산에저 140
- 건강검진자를대상으로해석적보고를위한전문가시스템의개발 - Fig. 2. Screen showing selection of test items for interpretative report. Fig. 3. Screen showing results interpreted by knowledge based expert system. 141
- 이채훈 - Table 3. List of active expert rules according to test results of subjects Active expert rule definition Number of subjects % of total subjects Urinalysis ph >8.0 1 0.2 ph <4.5 0 0 Bilirubin>(±), serum total bilirubin >1.5 mg/dl 2 0.4 Urobilinogen >(±) 0 0.0 Urobilinogen <(±) 0 0.0 Glucose >(±) 3 0.6 protein >(±) 10 1.9 protein =(±) 1 0.2 Nitrite >(±) 0 0.0 leukocyte >(±), WBC >5/HPF 17 3.2 Ketone >(±), normal serum glucose level 2 0.4 Blood >(±), RBC <5/HPF 13 2.4 Blood >(±), RBC >5/HPF 4 0.7 Anemia Hb <12.5 g/dl, normal MCV 37 6.9 Hb <12.5 g/dl, MCV <80 fl 6 1.1 Hb <12.5 g/dl, MCV >100 fl 0 0.0 White Blood Cells in Blood >10.0/mm 3 7 1.3 <4.0/mm 3 10 1.9 Thyroid function screening TSH >4.0 µiu/ml, normal Free T4 and Free T3 level 31 5.8 TSH >4.0 µiu/ml, Free T4 <0.9 ng/dl 2 0.4 TSH <0.23 µiu/ml, Free T4 >2.0 ng/dl 4 0.7 TSH <0.23 µiu/ml, normal Free T4 level 4 0.7 LIpid profile TG >160 mg/dl, Cholesterol <200 mg/dl glucose >110 mg/dl 5 0.9 Triglyceride >800 mg/dl 2 0.4 High-density lipoprotein cholesterol >60 mg/dl 36 6.7 Total Cholesterol >200 mg/dl 182 34.0 Tumor maker PSA >3.0 ng/ml 8 1.5 CEA >10.0 ng/ml 2 0.4 CA125 >35 U/mL 7 1.3 AFP >20.0 mg/dl 1 0.2 CA19-9 >37 U/mL 6 1.1 HBV Maker HBsAg(-), Anti-HBs(+), Anti-HBc(+) 13 2.4 HBsAg(-), Anti-HBs(+), Anti-HBc(-) 6 1.1 HBsAg(+), Anti-HBs(-), Anti-HBc(+), normal ALT level 19 3.6 HBsAg(+), Anti-HBs(-), Anti-HBc(+), abnormal ALT level 6 1.1 HBsAg(-), Anti-HBs(-), Anti-HBc(+) 9 1.7 VDRL&TPHA VDRL(+) or (±), TPHA(+) or (±) 2 0.4 VDRL(-), TPHA(+) or (±) 4 0.7 Other infectious Condition Anti-HCV(+) or (±) 3 0.6 Rheumatoid factor >20 IU/mL 15 2.8 Eosinophil in blood >8 % 17 3.2 ASO >200 IU/mL 10 1.9 C-reative protein >0.7 mg/dl 12 2.2 Diabetes Glucose >120 mg/dl 23 4.3 HbA1c >7.0 % 26 4.9 Other6 amylase >150 IU/L 12 2.2 calcium <8.4 mg/dl 3 0.6 calcium >10.6 mg/dl 1 0.2 bilirubin >2.0 mg/dl 0 0.0 Creatinine >1.7 mg/dl and BUN >23 mg/dl 1 0.2 Creatine Kinase >230 U/L 18 3.4 GGT >60 IU/L 53 9.9 ALT >45 IU/L, AST >40 IU/L, GGT >60 IU/L 8 1.5 Alkaline phosphatase >270 IU/L, normal GGT 3 0.6 uric acid >7.8 mg/dl 9 1.7 Abbreviation:Hb, Hemoglobin; MCV, Mean corpuscular volume ; GGT, γ-glutamyl transpeptidase; AST, Aspratate Aminotransferase; ALT, Alanine Aminotransferase; BUN, Blood Urea Nitrogen; CA19-9, carbohydrate antigen 19-9; AFP, Alpha-fetoprotein; CA125, carbohydrate antigen 125; CEA, Carcinoembryonic antigen; PSA, Prostate Specific Antigen; RBC, Red Blood Cell; HbA1c, glycosylated haemoglobin; VDRL, Venereal Disease Research Laboratory; TPHA, Treponema pallidum haemagglutination; TSH, thyroid stimulating hormone; Free T4, free thyroxine; Free T3, serum free triiodothyronine. 142
- 건강검진자를대상으로해석적보고를위한전문가시스템의개발 - 장되어있는검체번호를일차적으로개인용컴퓨터에출력하고, 각검체번호를선택시검사항목과결과를보여줄수있도록하였으며, 전문가시스템사용은검진일에실시한검사항목들을모두선택한후실행하면검사결과에부합되는부가설명이출력되게하였다 (Fig. 2, Fig. 3). 그리고, 최종보고서는 Table 2와같은형태로구성하였다. 개발환경은 Window XP (microsoft, Seattle, USA) 에서 Delphi 5.0(Borland Corp. Scotts Valley CA, USA) 를이용하여개발하였다. 2. 프로그램의검증및대상자 2007년 8월영남대학교병원건강증진센터를방문한 535명 ( 남자 : 284명, 여자 : 251명, 연령 : 47.5±10.9(17-75)) 을대상으로진단검사의학과에의뢰된검사를이용하여개발한전문가시스템의유용성을검증하였다. 결과대상기간중전체검진자중 49.5%(265/535) 에서진단검사의학과에서의뢰된검사에서정상적인검사결과를보였고, 50.5%(270/535) 에서전문가시스템에서 1개이상의경고치를나타내었다 (Table 3). 가장많은경고는고지혈증으로 34%(182/535) 에서총 cholesterol 이 200 mg/dl이상으로나타났다. 소변검사에서는 9.9% (53/535) 에서경고가생겼으며, dip stick과현미경상에백혈구의증가로인한요로감염이의심되는경우가 3.2%(17/535), 적혈구검출로인한경고가 2.4%(13/535) 에서나타났다. 빈혈지표에서는 8%(43/535) 에서경고가나타났으며, 대부분 (6.9%, 37/535) 은정적혈구정색소성 빈혈로나타났으나, 용혈성빈혈의검사소견에선경고치가발생하지않아초기철결핍성빈혈이경우로생각되었으나, 모든검진자에서는추적조사가시행하지는못하였다. 저색소성소구성빈혈은 6례 (1.1%, 6/535) 에서나타났으나, 고혈색소성빈혈은검출되지않았다. 1.3% (7/535) 에서백혈구증가증이나타났으며, 1.9% (10/535) 에서백혈구감소증이나타났다. 갑상선기능검사에서 thyroid stimulating hormone (TSH) 의증가는있으나, 유리 thyroxine(free T4) 와유리 triiodothyronine(free T3) 가정상수치를보인경우가 5.8%(31/535) 에서나타났다. 또한, 갑상성기능저하증이의심되는경우가 2례 (0.4%, 2/535), 갑상선기능항진증이의심되는경우가 4례 (0.7%, 4/535) 에서관찰되었다. 고지혈관련지표에서는즉각적인치료가필요한기준으로정한 800 mg/dl이상의 triglyceride 가 2례에서관찰되었다. 암표지자검사에서 24례 (4.5%, 24/535) 에서경고치가발생하였고, prostate specific antigen(psa) 8례, carbohydrate antigen 125(CA125) 7례, carbohydrate antigen 19-9 (CA19-9) 6례에서증가되어대부분을차지하였고, 추가적인추적조사가필요하였다. 그리고, hepatitis B virus antigen(hbsag) 가양성인경우가 25례 (4.7%, 25/535) 가나타났으며, 그중 alanine aminotransferase(alt) 가증가된경우가 6례에서만나타났다. 매독항원검사에서이상이보인경우는 1.1%(6/535) 였고, 그외혈중 glucose의증가가 24.3%(23/535) 에서관찰되었고, 담관폐쇄와알콜성간염과연관이있는 γ-glutamyl transpeptidased(ggt) 의단독증가가 9.9%(53/535) 에서나타났으며, ALT, aspratate aminotransferase(ast) 의증가를보여간기능에이상이나타난경우는 1.5%(8/535) 143
- 이채훈 - 였으며, 이경우 HBsAg 양성이 4례로나타났다. 이차정밀검진은전체검진대상자중 8.2% (49/535) 에서시행되었으며, 경고치가나타난대상자중 18.1%(49/270) 에서시행되었다. 고 진단검사의학과의전산화는검사처리의효율성을증가시켜빠른검사결과의보고, 조회및관리에이용되고있으며, 16-21) 최근에는컴퓨터를이용하여일부검사항목에대해전문가 5, 6, 10, 22) 시스템을이용한결과판독및해석적보고의시도가점차적으로증가되고있지만아 8, 23-26) 직까지초보적인단계이다. 이러한해석적보고의시도가증가되는이유는검사의결과값만을보고하는것보다임상의가치료, 진단및경과추적에더많은정 1, 8, 11) 보를제공할수때문이다. 즉과거 10년동안생체의학 (biomedical science) 의발전으로환자의진단, 치료및연구에이용되고있는생화학및질환의유전적표지자등의검사항목이증가하여, 정밀진단을위한이러한검사선택과해석에대한요구가필요하게되었으 27, 28) 며, 미국에서과거몇년동안검사의선택과해석에관한임상의요구가증가되고있다. 29) 또한부적절한검사의사용은진료비의증가, 잘못된의학적결과 (medical outcomes) 을초래할수있다. 30) 따라서, 해석적보고는임상의에게결과해석의오류를감소시키고 [24], 추가적검사를선택및처방할수있도록하는 1, 29) 것이중요한목적이다. 이러한점은정도관리등검사실질적관리만시행하고검사결과만을보고하는경우얻을수없는장점이다. Kratz 등 24) 은검사항목에대한충분한검사지 찰 식이없는소규모병원과개업의에게는해석적보고가많은도움이되며, 특히항핵항체, cardiac enzyme indices, 자주처방하지않는검사에대해해석적보고는반드시필요하다고했다. 또한 Smith 등 11) 는부분적이지만인공지능시스템을이용한해석적보고는진단시간및경비의감소시키는효과가있었다고하였다. 해석적보고시고려해야할사항은검사항목의감수성과특이성뿐만아니라, 측정전, 측정중에생길수있는오류, 검사시간섭현상등의분석과정, 및검사항목의임상적의의에대한충분히이해등이지만, 가장큰문제점은환자의임상적양상에대해알아야한다는점 1, 7, 9, 23) 이다. 병리과와영상의학과의해석적보고가진단검사의학과분야에서도특수검사항목의수적증가와복잡성으로인해점차적으로증가될것으로추측되고있지만, 29) 진단검사의학과검사의판독은해부학적인자료만으로판독이가능한병리과와영상의학과와달리임상적정보없는검사해석은잘못된정보를제공할 9, 24) 수있으므로판독에한계가있는것도사실이다. 이러한판독오류를감소시키기위해최근에는영국과호주등에서외부정도관리에서도결과해석에대한정도관리를시도하고있 1, 26) 다. 또한해석적보고는검사결과가측정장비에서출력된후판독을위해일정시간이소요되므로, 최종보고가지연되어환자의치료에문제점을야기할수도있다. 9) 따라서최종적결과보고시간을감소시키기위해전문가시스템을이용하여일차적인판독를실시한후보고하는것도좋은방법이라고생각된다. 본연구는향후입원및외래에내원한환자를대상으로해석적보고를위해기초자료를확보하기위해실시한것이다. 일부에선환자 144
- 건강검진자를대상으로해석적보고를위한전문가시스템의개발 - 의충분한임상적정보없이전산등을이용하여일률적이고고정된형태의보고의필요성에 9, 24) 부정적의견도있으나, 저자는입원초기혹은초진시선별검사의경우에는전산화된보고도임상에도움이될수있을것으로생각한다. 해석적방법으로위에서기술한전산등을이용한고정된형태의보고이외에환자개개인에따라달리보고하는방법이있으며, 임상의는일정한양식의해석적보고가아닌환자에따라달리해석한보고를원한다. 24) 본연구는 rule-based expert system를이용하여각검사항목의이상치를전산으로해석한고정된형태의보고이며, 증례의연구와추가적인해석보고양식을보완하여환자개개인에따른보고형태가가능하도록하여야할것으로사료된다. 본연구결과에서살펴보면식생활의변화등으로고지혈증의빈도가 34% 에서나타나이에대한예방과치료에사회적인관심이필요해보인다. 또한 HBsAg양성인경우가 25예에서나타났으나 21예에서 ALT 등간기능지표에서정상으로보여 B형간염바이러스보균자가많은것으로보였지만이는병원내환자의분포와는다를것으로생각되며추가적룰이필요한부분이다. 해석적보고시고려해보아야할점은검사결과의지나친확대해석 (over-interpretation) 이다. 25) 본연구에서도대상자의 50.5% 에서경고치가발생하였지만, 진단검사의학과검사와관련된이차정밀검사는 18.1%(49/270) 에서만실시한것으로조사되어확대해석의여지가있는것으로생각된다. 환자의임상적정보없는결과판독은환자상태를과대혹은과소평가를하게되지만, 간과된감별질환을한번더생각해봄으로써환자진료에도움이될것 으로생각되며, 임상의는임상적소견이나다른검사소견들을검토하여종합적으로해석하는것이바람직할것이다. 또한건강검진자의특성상질환의분포가다양하지못한것이연구기간중아쉬움으로나타났으며, 임상적용시추가적인룰의계속적인보완이필요할것으로생각된다. 결론적으로종합적해석및임상적적용능력을기르기위한노력이필요하고, 앞으로많은경험이축적된다면앞서말한몇가지제한점들을해결하고더욱흥미로운결과를얻을수있을것으로보인다. 그리고, 해석적보고는입원환자의종합적인검증및판독보고서를작성한경우에임상병리검사종함검증료가산정되므로병원에경제적으로기여할수있고, 환자의진료에있어서매우유용한정보를제공할수있을것으로생각된다. 요약배경 : 검사결과의판독과보고는진단검사의학과에서시행해야할업무이지만일일발생하는검사건수의방대함으로쉽게접근하기어려움이있어검사결과의판독과해석을할수있는보조프로그램의개발이필요하다. 따라서저자는판독프로그램을개발하여검사항목이적으며, 묶음검사항목으로되어있는건강검진자를대상으로임상적용시의문제점과보완점을알아보고자하였다. 재료및방법 : 연관성이있는검사항목을기준으로각검사결과에따른 rule-based expert 시스템을개발하여 2007년 8월본원건강증진센터를방문한 535명을대상으로검증하였다. 룰의구성은 if-then 으로구성하였으며, 즉 145
- 이채훈 - 검사결과가조건에맞는경우미리입력되어있는판독문구가출력되도록하였다. 결과 :50.5% 에서해석적보고가필요한것으로나타났다. 고지혈증이 34%, 빈혈관련질환이 8.0%, 뇨검사에서 9.9%, 종양표지자가 4.7%, 갑상선장애가 7.7%, B형간염표지자에서 4.7%, 혈당관련질환에서 4.3%, 등으로나타났으며, 이는실제추적검사를시행한환자수 (49명, 9.1%) 보다상당히높게나타났다. 이는해석적보고가단순히검사항목에대한검증으로대상자의임상적소견을감안하지않아더높은비율로나타난것으로보인다. 결론 : 임상의에보고하기전에시행하는검사의판독은검사실내의정도관리적측면에서도도움이되며, 임상의가최종진단전고려해야할질환등을감별함으로써보다질적인측면에서향상된정보를제공함으로써입원및외래내원환자에적용하면치료등에도움이될수있을것으로생각된다. 참고문헌 1. Lim EM, Sikaris KA, Gill J, Calleja J, Hickman PE, Beilby J, et al. Quality assessment of interpretative commenting in clinical chemistry. Clin Chem 2004 Mar;50(3):632-7. 2. Lee SY, Kim JW, Jeon SS. Interpretative reporting system of the analysis of urine stone risk. Korean J Clin Pathol 1999 Dec;19 (6):629-36. 3. Ivandic M, Hofmann W, Guder WG. The use of knowledge-based systems to improve medical knowledge about urine analysis. Clin Chim Acta 2000 Jul;297(1-2):251-60. 4. Valdiguie PM. Role and use of expert systems within the clinical laboratory. Clin Chim Acta 1998 Dec;278(2):193-202. 5. Lee CH, Cho HS. Interpretation of antimirobial susceptibility test of Enterobacteriaceae to beta-lactams with expert system. Korean J Lab Med 2004 Dec;24(6):377-85. 6. Kim MH, Lee CK, Lee KN, Jeong BY. A study on the need of clinical expert system for acid-base diagnosis. Korean J Clin Pathol 1999 Oct;19(5):587-91. 7. Zeng Q, Cimino JJ, Zou KH. Providing conceptoriented views for clinical data using a knowledge-based system: an evaluation. J Am Med Inform Assoc 2002 May-Jun;9(3):294-305. 8. Burke MD. Clinical laboratory consultation: appropriateness to laboratory medicine. Clin Chim Acta 2003 Jul;333(2):125-9. 9. Young IS. Interpretive comments on clinical biochemistry reports. J Clin Pathol 2005 Jun; 58(6):575. 10. Trendelenburg C, Colhoun O, Wormek A, Massey KL. Knowledge-based test result in interpretation in laboratory medicine. Clin Chim Acta 1998 Dec;278(2):229-42. 11. Smith BJ, McNeely MD. The influence of an expert system for test ordering and interpretation on laboratory investigations. Clin Chem 1999 Aug;45(8 Pt 1):1168-75. 12. Freidman RB, Young DS. Effects of disease on clinical laboratory tests 3rd ed. Washington, DC: AACC Press; 1997. 13. Tietz NW. Clinical guide to laboratory tests. 3rd ed. Philadephia, Pennsylvania: W.B. Saunders Company; 1995. 14. McPherson RA, Pincus MR. Henry s clinical diagnosis and management by laboratory methods. 21st ed. Philadelphia, PA: Saunders Inc; 2007. 15. Suh JS, Kwon SW, et al. eds. Laboratory Medicine 3rd ed. Seoul, Korea: Medical Publishing Co.; 2001. 146
- 건강검진자를대상으로해석적보고를위한전문가시스템의개발 - 16. Wycoff DA, Wagner JR. Distributed laboratory computing: integration of a laboratory computer into a hospital information system. Am J Clin Pathol 1978 Sep;70(3):390-9. 17. Elevitch FR. Linking laboratory and hospital information systems: the modular approach. Clin Lab Med 1983 Mar;3(1):219-31. 18. Lee KN. Korea university laboratory data management and communication system(v). Korean J Clin Pathol 1992 Jun;23(2):159-63. 19. Kim JQ, Min WK, Lee SK, Cho HI, Park MH, Kin SI. Seoul national university clinical laboratory information system(i). Korean J Clin Pathol 1987 Jun;7(2):223-30. 20. Yoon JH, Cho HI. Lan based hospital information system (II) - Computerization of blood collection room to reduce clinical order turnaround time-. Korean J Clin Pathol 1998 Mar;18(2):288-92. 21. Park AJ, Kim HR, Lee MK. Networking experience of point-of-care test glucometer. Korean J Lab Med 2006 Aug;26(4):294-8. 23. Kratz A, Soderberg BL, Szczepiorkowski ZM, Dighe AS, Versalovic J, Laposata M. The generation of narrative interpretations in laboratory medicine: a description of servicespecific sign-out rounds. Am J Clin Pathol 2001 Dec;116 Suppl:S133-40 24. Kratz A, Laposata M. Enhanced clinical consulting-moving toward the core competencies of laboratory professionals. Clin Chim Acta 2002 May;319(2):117-25. 25. Lim EM, Vasikaran SD, Gill J, Calleja J, Hickman PE, Beilby J, Penberthy L, Sikaris KA. A discussion of cases in the 2001 RCPA- AQAP Chemical Pathology Case Report Comments Program. Pathology 2003 Apr;35 (2):145-50. 26. Sciacovelli L, Zardo L, Secchiero S, Zaninotto M, Plebani M. Interpretative comments and reference ranges in EQA programs as a tool for improving laboratory appropriateness and effectiveness. Clin Chim Acta 2003 Jul;333 (2):209-19. 22. Ivandic M, Hofmann W, Guder WG. Development and evaluation of a urine protein expert system. Clin Chem. 1996 Aug;42(8 Pt 1):1214-22. 27. Burke MD. Laboratory medicine in the 21st Century. Am J Clin Pathol 2000 Dec;114(6): 841-6. 28. Burke MD. Clinical laboratory consultation. Clin Chem 1995 Aug;41(8 Pt 2):1237-40. 29. Dighe AS, Soderberg BL, Laposata M. Narrative interpretations for clinical laboratory evaluations: an overview. Am J Clin Pathol 2001 Dec;116 Suppl:S123-8. 30. van Walraven C, Naylor CD. Do we know what inappropriate laboratory utilization is? A systematic review of laboratory clinical audits. JAMA 1998 Aug;280(6):550-8. 147