Insight Extraction From E-Health Bookings by Means of Hypergraph and Machine Learning.

New technologies are transforming medicine, and this revolution starts with data. Usually, health services within public healthcare systems are accessed through a booking centre managed by local health authorities and controlled by the regional government. In this perspective, structuring e-health data through a Knowledge Graph (KG) approach can provide a feasible method to quickly and simply organize data and/or retrieve new information. Starting from raw health bookings data from the public healthcare system in Italy, a KG method is presented to support e-health services through the extraction of medical knowledge and novel insights. By exploiting graph embedding which arranges the various attributes of the entities into the same vector space, we are able to apply Machine Learning (ML) techniques to the embedded vectors. The findings suggest that KGs could be used to assess patients' medical booking patterns, either from unsupervised or supervised ML. In particular, the former can determine possible presence of hidden groups of entities that is not immediately available through the original legacy dataset structure. The latter, although the performance of the used algorithms is not very high, shows encouraging results in predicting a patient's likelihood to undergo a particular medical visit within a year. However, many technological advances remain to be made, especially in graph database technologies and graph embedding algorithms.

Predictive Medicine for Salivary Gland Tumours Identification Through Deep Learning.

Nowadays, predictive medicine begins to become a reality thanks to Artificial Intelligence (AI) which allows, through the processing of huge amounts of data, to identify correlations not perceptible to the human brain. The application of AI in predictive diagnostics is increasingly pervasive; through the use and interpretation of data, the first signs of some diseases (i.e. tumours) can be detected to help physicians make more accurate diagnoses to reduce the errors and develop methods for individualized medical treatment. In this perspective, salivary gland tumours (SGTs) are rare cancers with variable malignancy representing less than 1% of all cancer diagnoses and about 5% of head and neck cancers. The clinical management of SGTs is complicated by a high rate of preclinical diagnostic errors. Today, fine needle aspiration cytology (FNAC) represents the primary diagnostic tool in the hands of clinicians. However, it provides information that about 25% of cases are dubious or inconclusive, complicating therapeutic choices. Thus, finding new tools supporting clinicians to make the right choices in doubtful cases is necessary. This research work presents and discusses a Deep Learning-based framework for automatic segmentation and classification of salivary gland tumours. Furthermore, we propose an explainable segmentation learning approach supporting the effectiveness of the proposed framework through a per-epoch learning process analysis and the attention map mechanism. The proposed framework was evaluated with a collected CT dataset of patients with salivary gland tumours. Experimental results show that our methodology achieves significant scores on both segmentation and classification tasks.

Adnexal masses in pregnancy: fetomaternal blood flow indices during laparoscopic surgery.

STUDY OBJECTIVE: To assess changes in uterine and umbilical arteries during laparoscopy in human pregnancy. DESIGN: Case series (Canadian Task Force classification III). SETTING: University tertiary care referral center for high-risk pregnancy and minimally invasive surgery. PATIENTS: Nine pregnant women who underwent first- and second-trimester laparoscopic surgery because of an adnexal mass. INTERVENTION: Laparoscopic cyst enucleation or annessiectomy. MEASUREMENTS AND MAIN RESULTS: No maternal complications and no miscarriages or adverse pregnancy outcome occurred. Mean (SD) gestational age at delivery was 39.1 (0.7) weeks, birth weight was 3390 (298) g, and Apgar score at 5 minutes was 9.6 (0.5). Mean uterine resistance index, umbilical artery pulsatility index, and fetal heart rate were measured using transvaginal ultrasonography at various times during surgery. Mean uterine resistance index and umbilical artery pulsatility index values remained constant during laparoscopy. Fetal heart rate was maintained in the normal range (120-160 bpm) but progressively decreased during the surgical procedure. CONCLUSION: In human pregnancy, laparoscopic techniques do not seem to modify uteroplacental perfusion evaluated using noninvasive ultrasonography.

Laparoscopic versus vaginal hysterectomy for benign pathology.

PURPOSE OF REVIEW: The following review is an attempt to clarify the current role of vaginal and laparoscopic approaches in case of hysterectomy for benign pathologies. RECENT FINDINGS: Recent researches establish that vaginal hysterectomy seems to be the gold standard in case of benign pathologies and should be performed in preference to abdominal hysterectomy wherever possible. When vaginal hysterectomy is not technically possible, laparoscopic hysterectomy is to be preferred to abdominal hysterectomy. No advantages of laparoscopic hysterectomy could be found over vaginal hysterectomy, in particular, because laparoscopic hysterectomy is associated with a higher rate of complications (especially bladder and ureteral injuries). Other authors, instead, show that laparoscopic hysterectomy permits a safe bilateral salpingo-oophorectomy (BSO), the treatment of additional pathologies at the time of surgery, a reduction of intraoperative bleeding, postoperative pelvic pain and length of stay compared with vaginal hysterectomy. Moreover, laparoscopic approach, in experienced hands, is not associated with any increase in major complication rate. SUMMARY: It's time to reduce abdominal hysterectomy in favour of mini-invasive approaches. To do so a suitable training and supervision are paramount before embarking on total laparoscopic hysterectomy (so that complications are minimized) or on vaginal hysterectomy (so to perform a planned BSO). It is also essential, particularly for total laparoscopic hysterectomy, to share the techniques used by different surgeons, the results and the complications concerning this approach.

Laparoscopic vs vaginal hysterectomy for benign pathology.

OBJECTIVE: The objective of the study was to compare length of stay, blood loss, operative time, and pain of laparoscopic and vaginal hysterectomy. STUDY DESIGN: This was a prospective, randomized, controlled comparison between vaginal (VH) and laparoscopic (LH) hysterectomy among 60 consecutive patients with a uterine volume of 300 mL or less and without uterine prolapse. Patients were followed up for 12 months. RESULTS: The groups were significantly different for mean operative time (VH: 81 +/- 30 minutes; LH: 99 +/- 25 minutes; P = .033) and blood loss (LH: 83 +/- 57 mL; VH: 178 +/- 149 mL; P = .004). Bilateral adnexectomy was performed when preoperatively planned in 73% of cases of the vaginal arm, whereas it was always performed in the laparoscopic arm (P = .045). Postoperative pain on day 0 and the number of days of analgesic request were higher in the vaginal group (P = .023 and P = .017, respectively). LH was associated with a reduced hospital stay (LH: 2.7 +/- 0.5 days; VH: 3.2 +/- 0.6 days; P < .001).There were no differences between the groups at the follow-up. CONCLUSION: Laparoscopic hysterectomy results in a shorter hospital stay, less blood loss, and less postoperative pain compared with vaginal hysterectomy.

Postsurgical ovarian failure after laparoscopic excision of bilateral endometriomas.

OBJECTIVE: This study was undertaken to determine the frequency of postsurgical ovarian failure in patients undergoing laparoscopic excision of bilateral endometriomas. STUDY DESIGN: Patients who had been operated on for bilateral ovarian endometriosis between January 1995 and December 2003 and who were younger than 40 years at the time of surgery were contacted by telephone and interviewed. RESULTS: A total of 126 patients were recruited. Mean +/- SD age of patients at the time of surgery was 30.4 +/- 4.3 years. Postsurgical ovarian failure was documented in 3 cases, corresponding to a rate of 2.4% (95% CI 0.5%-6.8%). In all cases, this complication occurred immediately after surgery. CONCLUSION: Patients who had been operated on for bilateral endometriomas have a low but definite risk of premature ovarian failure occurring immediately after surgery.

Human chorionic gonadotropin level as a predictor of trophoblastic infiltration into the tubal wall in ectopic pregnancy: a blinded study.

OBJECTIVE: To determine the relationship between gestational age, tubal ultrasonographic diameter, and serum hCG levels and different stages of trophoblastic infiltration of the tubal wall in ectopic pregnancy. DESIGN: Blinded prospective study. SETTING: University-based clinic in Italy. PATIENT(S): Thirty-seven consecutive patients with an ampullary ectopic pregnancy. INTERVENTION(S): Laparoscopic salpingectomy. MAIN OUTCOME MEASURE(S): Gestational age, diameter of the tubal mass as determined by transvaginal ultrasonography. and hCG level on the day of surgery. Ectopic pregnancy was classified according to the depth of trophoblastic infiltration: trophoblast limited to the tubal mucosa (stage I), extension to the tubal muscularis (stage II), or complete tubal wall infiltration up to the serosa discontinued by trophoblastic cells (stage III). RESULT(S): Fifteen patients (40.5%) had stage I tubal infiltration, 14 (37.8%) had stage II infiltration, and 8 (21.6%) had stage III infiltration. Gestational age and diameter of the tube did not differ among the three groups. The median hCG level was 1,710.5 mIU/mL (range, 113-5,635 mIU/mL) for patients with stage I infiltration. 4,690.0 mIU/mL (range, 150-21,531 mIU/mL) for patients with stage II infiltration, and 15,700.0 mIU/mL (range, 13,809-21,650 mIU/mL) for patients with stage III infiltration. All the patients with hCG levels > 6,000 mIU/mL had stage II or III invasion. CONCLUSION(S): These findings may explain why the conservative treatment of ectopic pregnancy is less successful in patients with high hCG levels than in patients with low levels. Use of radical procedures may be justified in the former group.

Human chorionic gonadotropin patterns after a single dose of methotrexate for ectopic pregnancy.

OBJECTIVE: The great variability in human chorionic gonadotropin (HCG) levels after a single dose of methotrexate (MTX) for ectopic pregnancy makes it difficult to predict treatment failure. We describe different patterns of HCG levels. STUDY DESIGN: Fifty patients were injected i.m. with 50mg/m(2) of MTX for an ectopic pregnancy. Venous blood samples for HCG detection were obtained on the day of treatment (day 0), day 3 and day 7 and weekly until values were undetectable. Patients were classified as: group 1, persistent pathology (n=11); group 2, complete resolution with a decrease of HCG levels at day 3 (n=30); group 3, complete resolution after a rise of HCG values at day 3 (n=9). Statistical analysis was performed using the Mann-Whitney non-parametric test with 95% confidence intervals. RESULTS: Values of day 0 were similar for all the groups. HCG levels of group 3 decreased rapidly after day 3 and at day 7 they were significantly different from levels of group 1. Differences in HCG levels between groups 2 and 3 became indistinguishable from day 21. CONCLUSION: The observation of patients undergoing resolution after an initial increase of HCG levels justify an expectant management for 1 week in clinically stable patients. The strategy to separate HCG curves in patients undergoing resolution may shed light on the different clinical responses to therapy for ectopic pregnancies. However, the phenomenon of the immediate rise of HCG should be better investigated.