Patients classification on weaning trials using neural networks and wavelet transform.

The determination of the optimal time of the patients in weaning trial process from mechanical ventilation, between patients capable of maintaining spontaneous breathing and patients that fail to maintain spontaneous breathing, is a very important task in intensive care unit. Wavelet Transform (WT) and Neural Networks (NN) techniques were applied in order to develop a classifier for the study of patients on weaning trial process. The respiratory pattern of each patient was characterized through different time series. Genetic Algorithms (GA) and Forward Selection were used as feature selection techniques. A classification performance of 77.00±0.06% of well classified patients, was obtained using a NN and GA combination, with only 6 variables of the 14 initials.

Primary tracheal lymphoma causing respiratory failure.

Lymphomatous involvement of the airway causing stridor is a rare but frightening presentation of an eminently treatable condition. We describe a 24-year-old woman with tracheal non-Hodgkin lymphoma who was initially diagnosed with asthma, but subsequently presented with near-fatal acute upper airway obstruction because of a tracheal Anaplastic Lymphoma Kinase (ALK)+ anaplastic T-cell lymphoma. The obstructing tumor was extricated by means of rigid bronchoscopy. After six cycles of Cyclophosphamide, Doxorubicin, Vincristine, Prednisolone chemotherapy, the patient went into complete clinical remission. A high index of suspicion in patients with dyspnoea and wheeze unresponsive to bronchodilators is crucial in early diagnosis of tracheal tumors.

Curative irradiation of limited endobronchial carcinomas with high-dose rate brachytherapy. Results of a pilot study.

OBJECTIVES OF THE STUDY: Pilot study to assess high-dose rate (HDR) brachytherapy as sole treatment for limited endobronchial non-small cell lung carcinomas. INCLUSION CRITERIA: Proximal non-small cell lung cancer in a not previously irradiated area, with a maximal diameter of 1 cm, no visible tumor on CT scan, lack of other treatment options in patients with severe, chronic respiratory failure, surgery, or external radiotherapy for a previous lung cancer. TREATMENT PROTOCOL: Treatment was based on an escalating dose protocol. Patients received three to five fractions of 7 Gy prescribed at 10 mm from the center of the applicator, once a week. RESULTS: Nineteen patients were included in this trial. The first two patients received three fractions of 7 Gy, the four next patients received four fractions, and the 13 remaining patients were treated with five fractions of 7 Gy. Two months after the end of the procedure, tumors in 15 of 18 evaluable patients (83%) were locally controlled with negative results of biopsies. At 1 year, local control was still obtained in 12 of 16 evaluable patients (75%). With a mean follow-up of 28-months, 1-year and 2-year actuarial survival rates were 78% and 58%, respectively, with a 28-month median survival. One patient with local control died from hemoptysis 12 months after treatment. Two patients suffered from severe necrosis of the bronchial wall; one of them died from hemoptysis. CONCLUSIONS: HDR brachytherapy is an effective treatment for small endobronchial tumors. Late toxicity on the bronchial wall is still too high and was attributed mainly to contact between the catheter and the bronchial mucosa. Exclusive HDR brachytherapy should be restricted to carefully selected patients for whom there is no alternative curative treatment. New bronchial applicators and a lower dose per fraction may reduce the incidence and attenuate the severity of late complications.