Study Objective 1) Demonstrate non-invasive cutaneous Electroviscerography (EVG)-detected abnormal gastrointestinal myoelectrical activity (GMA) has 100% accuracy diagnosing endometriosis. 2) Demonstrate GMA-based artificial intelligence (AI)-derived formulae detects endometriosis regardless of hormonal suppression, surgical stage or symptoms. Design Prospective open-labeled study. Setting Clinical center for women's healthcare. Patients or Participants 50 women ages 17-45 with chronic abdominal pain, negative CT or MRI, and gastrointestinal endoscopy, scheduled for laparoscopy. Interventions Baseline EVG w/water load satiety test(WLST) and pain scores, followed by diagnostic laparoscopic surgery w/biopsy. Measurements and Main Results Response to EVG WLST identified specifically elevated GMA frequencies of 15-20(9.5 vs 2.2, p<0.001), 30-40(3.1 vs 0.8, p<0.001) and 50-60(1.1 vs 0.5,p<0.05) compared to known population normals and disease free subjects with abdominal pain. GMA-derived 2 stage Ai-determined threshold (AIET) was calculated to be >0.5, using linear regression of area under curve on first pass and age and pain scores in the second pass, demonstrated 100% selectivity and 100% specificity, respectively. All subjects exceeded AIET and had ASRM stages 1-4 at laparoscopy. AIET scores ranged 0.52 to 8.9, and escalated according to advancing surgical grade. Hormonal suppression did not affect diagnosis. Conclusion Endometriosis-specific abnormalities of GMA post-WLST predicts surgically detected endometriosis w/100% accuracy. AI GMA-derived threshold detects endometriosis with 100% sensitivity and specificity. Hormonal suppression, symptoms and surgical grade of endometriosis did not affect test accuracy. Variations in Ai-derived threshold scores may predict surgical stage of endometriosis. EVG represents an ideal test for the detection and screening of endometriosis. 1) Demonstrate non-invasive cutaneous Electroviscerography (EVG)-detected abnormal gastrointestinal myoelectrical activity (GMA) has 100% accuracy diagnosing endometriosis. 2) Demonstrate GMA-based artificial intelligence (AI)-derived formulae detects endometriosis regardless of hormonal suppression, surgical stage or symptoms. Prospective open-labeled study. Clinical center for women's healthcare. 50 women ages 17-45 with chronic abdominal pain, negative CT or MRI, and gastrointestinal endoscopy, scheduled for laparoscopy. Baseline EVG w/water load satiety test(WLST) and pain scores, followed by diagnostic laparoscopic surgery w/biopsy. Response to EVG WLST identified specifically elevated GMA frequencies of 15-20(9.5 vs 2.2, p<0.001), 30-40(3.1 vs 0.8, p<0.001) and 50-60(1.1 vs 0.5,p<0.05) compared to known population normals and disease free subjects with abdominal pain. GMA-derived 2 stage Ai-determined threshold (AIET) was calculated to be >0.5, using linear regression of area under curve on first pass and age and pain scores in the second pass, demonstrated 100% selectivity and 100% specificity, respectively. All subjects exceeded AIET and had ASRM stages 1-4 at laparoscopy. AIET scores ranged 0.52 to 8.9, and escalated according to advancing surgical grade. Hormonal suppression did not affect diagnosis. Endometriosis-specific abnormalities of GMA post-WLST predicts surgically detected endometriosis w/100% accuracy. AI GMA-derived threshold detects endometriosis with 100% sensitivity and specificity. Hormonal suppression, symptoms and surgical grade of endometriosis did not affect test accuracy. Variations in Ai-derived threshold scores may predict surgical stage of endometriosis. EVG represents an ideal test for the detection and screening of endometriosis.