Staining, magnification, and algorithmic conditions for highly accurate cell detection and cell classification by deep learning.

OBJECTIVES: Research into cytodiagnosis has seen an active exploration of cell detection and classification using deep learning models. We aimed to clarify the challenges of magnification, staining methods, and false positives in creating general purpose deep learning-based cytology models. METHODS: Using 11 types of human cancer cell lines, we prepared Papanicolaou- and May-Grünwald-Giemsa (MGG)-stained specimens. We created deep learning models with different cell types, staining, and magnifications from each cell image using the You Only Look Once, version 8 (YOLOv8) algorithm. Detection and classification rates were calculated to compare the models. RESULTS: The classification rates of all the created models were over 95.9%. The highest detection rates of the Papanicolaou and MGG models were 92.3% and 91.3%, respectively. The highest detection rates of the object detection and instance segmentation models, which were 11 cell types with Papanicolaou staining, were 94.6% and 91.7%, respectively. CONCLUSIONS: We believe that the artificial intelligence technology of YOLOv8 has sufficient performance for applications in screening and cell classification in clinical settings. Conducting research to demonstrate the efficacy of YOLOv8 artificial intelligence technology on clinical specimens is crucial for overcoming the unique challenges associated with cytology.

Cardiopulmonary exercise testing for patients with anorexia nervosa: a case-control study.

PURPOSE: Patients with anorexia nervosa (AN) require appropriate nutrient therapy and physical activity management. Eating disorder treatment guidelines do not include safe, evidence-based intensity criteria for exercise. This study used cardiopulmonary exercise testing (CPX) to evaluate the exercise tolerance of patients with AN. METHODS: CPX was performed with 14 female patients with AN admitted to a specialized eating disorder unit between 2015 and 2019. Their anaerobic threshold (AT) was determined by assessing their exercise tolerance using CPX and compared with 14 healthy controls (HC). The metabolic equivalents (AT-METS) were compared when AT was reached. We examined factors related to AT (AN-AT) in the AN group, including age, body mass index (BMI), previous lowest weight, minimum BMI, past duration of BMI < 15, exercise history, and ΔHR (heart rate at the AT-resting heart rate). RESULTS: The AT of the AN group (BMI: 15.7 [Mean] ± 1.8 standard deviation [SD]) was significantly lower than that of the HC group (BMI: 19.7 ± 1.8) (AN: 10.0 ± 1.8 vs. HC: 15.2 ± 3.0 ml/kg/min, P < 0.001). AT-METS was also significantly lower in the AN group than in the HC group (AN: 2.9 ± 0.52 vs. HC: 4.4 ± 0.91, P < 0.001). AN-AT was highly influenced by ΔHR. CONCLUSIONS: This study showed that AT and AT-METS were lower in patients with AN than in HC. Patients with AN should be prescribed light-intensity aerobic exercise, and the current findings may help develop future physical management guidelines for patients with AN. LEVEL OF EVIDENCE: III: Evidence obtained from case-control analytic studies.

Potential impact of the COVID-19 pandemic on japanese patients with eating disorders -a cross-sectional study.

BACKGROUND: The COVID-19 pandemic and associated lockdown had a considerable impact on eating disorders (EDs). We evaluated the clinical features of Japanese ED patients before and after the first COVID-19 outbreak-related state of emergency (April 7, 2020). METHODS: We studied 148 patients who were divided into two groups based on when they arrived at our clinic: before (Before group: n = 86) or after (After group: n = 62) the start of the first state of emergency. All patients completed the Japanese versions of the Eating Disorder Inventory (EDI) and Parental Bonding Instrument (PBI). RESULTS: The After group was substantially younger than the Before group (p = .0187). Regardless of the ED type, patients who developed an ED during the first state of emergency tended to be significantly younger than those who developed one before. Differences in EDI characteristics were observed between the two groups. The PBI care subscale was notably higher (p = .0177) in the After group. The PBI maternal care subscale was the only statistically significant factor associated with age (ß = -0.35, p < .0001). CONCLUSIONS: Home confinement associated with the COVID-19 pandemic and the ensuing increase in parent-child closeness may have influenced the decreased age of ED patients at their initial consultation. Treatment interventions should consider the differences in the clinical features of EDs.

Cognitive behavioral therapy with interoceptive exposure and complementary video materials for irritable bowel syndrome (IBS): protocol for a multicenter randomized controlled trial in Japan.

BACKGROUND: There is growing evidence of the treatment efficacy of cognitive behavioral therapy (CBT) for irritable bowel syndrome (IBS). CBT is recommended by several practice guidelines for patients with IBS if lifestyle advice or pharmacotherapy has been ineffective. Manual-based CBT using interoceptive exposure (IE), which focuses on the anxiety response to abdominal symptoms, has been reported to be more effective than other types of CBT. One flaw of CBT use in general practice is that it is time and effort consuming for therapists. Therefore, we developed a set of complementary video materials that include psycho-education and homework instructions for CBT patients, reducing time spent in face-to-face sessions while maintaining treatment effects. The purpose of this study is to examine the effects of CBT-IE with complementary video materials (CBT-IE-w/vid) in a multicenter randomized controlled trial (RCT). METHODS: This study will be a multicenter, parallel-design RCT. Participants diagnosed with IBS according to the Rome IV diagnostic criteria will be randomized to either the treatment as usual (TAU) group or the CBT-IE-w/vid + TAU group. CBT-IE-w/vid consists of 10 sessions (approximately 30 min face-to-face therapy + viewing a video prior to each session). Patients in the CBT-IE-w/vid group will be instructed to pre- view 3- to 13-min videos at home prior to each face-to-face therapy visit at a hospital. The primary outcome is the severity of IBS symptoms. All participants will be assessed at baseline, mid-treatment, post-treatment, and follow-up (3 months after post assessment). The sample will include 60 participants in each group. DISCUSSION: To our knowledge, this study will be the first RCT of manual-based CBT for IBS in Japan. By using psycho-educational video materials, the time and cost of therapy will be reduced. Manual based CBTs for IBS have not been widely adopted in Japan to date. If our CBT-IE-w/vid program is confirmed to be more effective than TAU, it will facilitate dissemination of cost-effective manual-based CBT in clinical settings. TRIAL REGISTRATION: The trial was registered to the University Hospital Medical Information Network Clinical Trial Registry: UMIN, No. UMIN000030620 (Date of registration: December 28, 2017).

Portal hypertension in prolonged anorexia nervosa with laxative abuse: A case report of three patients.

OBJECTIVE: There has been no report on portal hypertension related to anorexia nervosa (AN). METHOD: We describe three cases of portal hypertension manifesting with collateral circulation represented by gastroesophageal varices in prolonged AN with laxative abuse and self-vomiting. These women, in their 20s to 50s, were diagnosed as having AN binging and purging type (AN-BP) that included self-induced vomiting and abuse of irritating laxatives (more than 100 tablets daily). RESULTS: Case 1 showed prominent ascites and a gastro-renal shunt on computed tomography scanning. Case 2 showed gastroesophageal varices on endoscopic examination. Case 3 showed gastroesophageal varices on computed tomography scanning and endoscopic examination. We performed liver biopsies in all patients and found only slight pericellular fibrosis. Our patients showed typical symptoms of portal hypertension, although liver cirrhosis was not present. DISCUSSION: We speculated that abnormal eating and purging behaviors were involved in the development of portal hypertension. We hypothesized that long-term laxative abuse, dehydration, and abnormal eating behavior are involved in the development of portal hypertension, considering these were common features in our patients. Portal hypertension and gastroesophageal varices should be considered as one of the potentially existing complications in prolonged AN-BP with self-induced vomiting and abuse of irritating laxatives.

Physical and psychological aspects of anorexia nervosa based on duration of illness: a cross-sectional study.

BACKGROUND: We evaluated physical and psychological features of patients with anorexia nervosa (AN) who differed by duration of illness. METHODS: Data were obtained from 204 female patients with AN, divided into two groups based on illness duration: short-term illness duration (less than 5 years; n = 118); and long-term duration (5 years or more; n = 86). Physical parameters were measured using blood serum testing and psychological aspects were assessed using various instruments. RESULTS: A significantly higher proportion of restricting type AN was observed in the short-term group while the proportion of binge eating/purging type AN was higher in the long-term group. There was no difference in body mass index (BMI) between the groups. Serum total protein, albumin, potassium, chloride, and calcium in the long-term group were significantly lower than in the short-term group. Overall scores on the Eating Disorder Inventory as well as most of the subscales, except maturity fears, were higher in the long-term group than in the short-term group. The care subscale of the Parental Bonding Instrument (PBI) was lower in the long-term group than in the short-term group, while the overprotection subscale of the PBI was higher in the long-term group than in the short-term group. Results of a multiple regression analysis indicated that the overprotection subscale of the PBI was the only significant predictor of duration of illness. CONCLUSIONS: Duration of illness may be associated with physical and psychological features of AN; thus, adapting therapeutic approaches to illness duration might be necessary.