Development of automatic generation system for lung nodule finding descriptions.

Worldwide, lung cancer is the leading cause of cancer-related deaths. To manage lung nodules, radiologists observe computed tomography images, review various imaging findings, and record these in radiology reports. The report contents should be of high quality and uniform regardless of the radiologist. Here, we propose an artificial intelligence system that automatically generates descriptions related to lung nodules in computed tomography images. Our system consists of an image recognition method for extracting contents-namely, bronchopulmonary segments and nodule characteristics from images-and a natural language processing method to generate fluent descriptions. To verify our system's clinical usefulness, we conducted an experiment in which two radiologists created nodule descriptions of findings using our system. Through our system, the similarity of the described contents between the two radiologists (p = 0.001) and the comprehensiveness of the contents (p = 0.025) improved, while the accuracy did not significantly deteriorate (p = 0.484).

Local hyperthermia with built-in endoscopy for radioresistant cervical cancer: a case series.

We evaluated the efficacy and safety of an endoscope-embedded transvaginal laser hyperthermia system for superficial cervical cancer that remained in the cervix after radiotherapy. We developed an innovative endoscope-embedded hyperthermia system consisting of a diode laser device, a temperature control unit, an endoscope control unit, and a transvaginal probe. Superficial lesions of recurrent or residual cervical cancer on the uterine cervix or vaginal wall after radiotherapy were eligible for this study. A total of four cases of three patients were eligible for this treatment. Case 1: The post-chemoradiotherapy residual tumor of a patient with stage IIB squamous cell carcinoma of the cervix was treated with the device. Two months after the laser hyperthermia treatment, the tumor's disappearance was confirmed. Case 2: A post-hysterectomy persistent tumor on the vaginal stump of a patient with stage IIB adenocarcinoma of the cervix was subjected to the laser hyperthermia treatment. Two months after the treatment, the stump's cytology was false positive. Case 3: As in case 2, this patient's recurrence in the anterior vaginal wall was subjected to laser hyperthermia treatment, but the tumor's growth was not controlled. Case 4: A tumor at the vaginal margin was identified during a salvage hysterectomy in a patient with stage IIB squamous cell carcinoma of the cervix who underwent chemoradiotherapy. After laser hyperthermia treatment, the tumor's disappearance was confirmed. Our new endoscope-embedded laser hyperthermia system can be a candidate for treating residual superficial cervical cancer after radiotherapy by accurately capturing superficial lesions.

A graded prognostic model for patients surviving 3 years or more (GPM ≥ 3Ys) after stereotactic radiosurgery for brain metastasis.

BACKGROUND AND PURPOSE: As more cancer patients with brain metastases (BMs) are surviving longer due to recent advancements in various treatment modalities, we developed a grading system for stereotactic radiosurgery (SRS)-treated BM patients with long survival. This is a Graded Prognostic Model for Patients Surviving 3 Years or More (GPM ≥ 3Ys). MATERIALS AND METHODS: First, using clinical factor-survival time analysis of 3237 patients in whom gamma knife (GK) SRS was performed by the second author (test cohort), we developed the GPM ≥ 3Ys based on survival ≥3 years as the objective variable. The validity of this model was then tested using another series of 3317 patients independently undergoing GK SRS performed by the third author (verification cohort). Number of patients surviving 3 years or more were 289 (8.9%) and 348 (10.5%), respectively. RESULTS: Using the test series, among various pre-SRS clinical factors, noted below, five were shown to be highly correlated with survival of ≥3 years. Therefore, we assigned scores for these five factors, i.e., "tumor numbers 1/2-4/≥5 (score; 6/1/0)", "female/male (5/0)", "KPS ≥80%/<80% (5/0)", "primary cancers of breast/lung/gastrointestinal tract/other (score; 1/0/3/0)", "controlled primary cancer/not (8/0)" and "existing extra-cerebral metastases/not (5/0). Patients were categorized into four grades according to the sum of scores, i.e., 0-9, 10-19, 20-29 and 30-36. Post-SRS mean survival times (MSTs) differed significantly (p < 0.0001) with no overlapping of 95% confidence intervals (CIs) among the four grades. Also, in the verification series, MSTs differed significantly (p < 0.0001) with no overlapping of 95% CI among the four grades of the GPM ≥ 3Ys system. CONCLUSION: Although this was a retrospective study, the GPM ≥ 3Ys system was shown to be very useful to physicians selecting among more aggressive treatment modalities for patients in whom longer survival can be expected.